baseballlover312 wrote:
It's quite obvious that I'm one of the toxic managers due to my difference of opinions and attitude. I'm not really sure what I've done to deserve this.
It's not the difference of opinions. It's about the attitude and you are the one who opened the discussion about it.
I actually like you as a manager and a member of the forum quite a lot. I like the way you build your team with a nationality focus and I'd be sad to see you withdrawing from the game because of the changes in game play instead of trying to adapt to them.
I understand that following the season when your team is highly affected is not fun at all but implying that you are going to leave the game if we don't go back to PCM15 feels like blackmail to me and I do not like it (this is not addressed specifically to you. There are some other managers who have implied this as well).
I dislike your attitude about this because you seem to be stuck with the idea that going back to PCM15 is the only solution and it's the only thing you are willing to accept. You are quick to jump to possible explanations of what might be breaking the game in one way or another, while seeming to ignore the possible solutions.
You've claimed "It hasn't knocked a few riders down a notch, it's taken half the leaders in the DB out of the equation completely". This is simply not true and you are trying to make things deliberately worse than they are because you do not like the way the new game plays out.
You said you are interested in making the game work well from transfers to standing. Standings working well is quite subjective here. Because you take the rankings of PCM15 as those of working well. PCM18 produces different results than PCM15 in certain scenarios, just like PCM15 produced different results compared to its predecessors. Why should we take the results of PCM15 as the correct ones?
As far as I'm concerned this season would be the worse in this aspect since no one really knew what to expect. And going back to PCM15 would not only stuck us in a repetitive loop, but it would also deem this season as basically worthless with nothing productive coming out of it. Now we know more about what we can expect from the game and everyone has the chance to adjust in the off-season accordingly.
I haven't mentioned going back to PCM 15 as the only solution. The only time I suggested I might leave the game if things stayed as they are now was in my initial post, which I have since admitted was emotional hyperbole meant to spark discussion and let out my frustration. I'm sorry I framed my post that way. Even in that post, I didn't take an explicit stand on PCM 15, and I admitted I was being a sore loser. Since then, seven pages and almost two weeks of discussion have happened. Since that original post, I have not once even suggested moving back to PCM 15, especially as the only solution. Yet you're saying I'm deliberately spreading misinformation purely to push that point. It really feels like a straw man is being built against me here, which is hard not to take personally.
I'll admit I haven't been very helpful actually suggesting solutions. That is totally valid criticism and I own that. I will admit most of my thoughts are about how the in-game experience reflects forum game things, not how technical aspects of gameplay affect the AI. That is why most of my comments have been negative in regards to solutions. I feel I can speak about what I don't think would work based on my experience and beliefs, but I don't want to speculate where I've had no experience. I can't endorse solutions that involve altering the DB in major ways because that goes against what I see as fundamentally important in the game. I don't know about he intricacies of stage making, I only know the results I've seen actually playing the game when I comment (i.e. road width and sprint trains).
My comments were not made to make things seem worse than they are. I wasn't trying to guilt people into siding with me, or anything like that. They were just how I felt. I can only speak from my experience, just like anyone else. A lot of riders on my team have become useless this season, as have a lot of similar riders on other teams. Not just lower scoring, but complete non-factors. That's what I saw, so that's what I said. It may be short sighted and even wrong, but it was not meant to mislead people. And yes, I was being a sore loser in saying it, I admit that.
On standing predictability, the problem is not that certain results are "correct" because they value certain stats the "right way." They are correct because they put riders who are worth more in the market generally in better positions. That's how the game works and feeds competition. Obviously everything should not be perfectly formulaic, and again, I never said that it should. We need races and excitement and unpredictability at times. We also need managers to make mistakes and overpay (something I'm well versed in), or negotiate especially brilliantly. That's the whole point of the game. But in order for those situations to matter, the general principle - that a better team gets you better results - must stand at least to the extent that it allows the market to function properly.
This season has made that principle very difficult because the game has fundamentally changed what a better team looks like, and at times, it doesn't align with actual strength, especially with how our specific DB functions. Some people on here have suggested that an adjustment to PCM 18 should be to purposefully obtain second tier sprinters instead of the best or have poor flat domestiques. To me, that doesn't fit with how teams can actually be constructed and rewarded using PCM and our market. So, if true, it is to some extent (however small) game breaking.
Of course the market will adjust over time, and huge OVL formula changes will go a long way. That will help reconcile that value problem, which to me is most important. However, as long as previous wage has influence on demanded wage, this will not be solved in a single year under the current system. This will take several years to be weaned out because of how dramatic some of the changes will be. Then comes another issue. Can we make those sweeping overall changes and still attempt to fix the stages to help those riders? Or do we risk overcompensating and having an opposite problem on our hands? I do fear doing both could hurt our adjustment long if not properly balanced.
The other thing is, the results of this season do matter. Not only as a game test, but for their substance. Yes, everyone can adjust accordingly this offseason, but there will be huge carry over effects. Riders becoming useless won't just hurt people this season - they have cascading effects. For me personally, it contributes to relegating and losing big talents. That could set me back for the next half decade. The same can be said for several other managers. I own a huge part of my poor team building, absolutely, and I'm not pinning it all on the game. My team is crap, and it always was. But there is an element of frustration when it feels like not all managers were properly informed ahead of time, and that put us at a disadvantage in that team building process when dealing with the new game.
For older managers, my perspective may seem crazy because the game has changed in huge ways before. As someone who only started in 2014, the degree of changes that have happened this year are unprecedented since I started the game, and that has certainly colored my perspective. This is not like when we moved from PCM 13 to 14 or 15, and I don't think we should have treated it as such and continue to treat it as such. That's all I'm saying.
Anyway, to clear things up, I will say it now so we can move on. I will not quit MG no matter the decisions we make about this topic. I didn't mean to blackmail anyone, and if it seemed like I was, I am sorry.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
Okay, so I spent some time to play through a few different scenarios with the stage Tamijo made. Fair warning, this is a long post. I've shortened it a lot, so there's more if you have any questions of if you think something's missing.
The stage looks like this, with the final 10km straight and wide roads:
I used the teams that participated in the actual Barbados Cycling Festival. For the first three experimental scenarios, I altered the original startlist a little, the final scenario is simply the actual startlist. I played through every scenario 10 times (except for one), there are screenshots for the first three to illustrate the way the sprints unfolded, then only the results for the other seven. I started a new one-off race every time to negate possible boni from yellow jerseys and "reset" possible favorite lists the game sets within each stage race. Here are the sprint leaders:
Spoiler
SP
AC
FL
RS
Coquard
84
83
73
74
Swift
84
82
74
70
Degenkolb
84
81
76
71
Ahlstrand
83
82
74
73
Ewan
83
81
75
71
Grosu
83
80
73
70
Avelino
83
78
71
66
Guerao
83
76
74
64
Kennaugh
82
81
77
71
Groenewegen
82
81
73
71
Cavendish
82
80
72
70
Vanderbiest
82
77
74
69
Boeckmans
81
83
73
71
Zabel
81
79
74
70
Appollonio
81
78
74
66
Vantomme
81
76
72
63
Itami
80
82
75
67
Guardini
80
82
72
69
Kip
80
82
69
63
Kupfernagel
80
81
78
71
Salleh
80
80
70
65
Nizzolo
80
79
73
67
Afewerki
79
80
73
70
For each of the first scenarios, I'm going to do a quick statistical overview for the Top 5, Top 10 and Top 15 sprinters. To make those categories, I just went from the top down in this list (So Top 5 is Coquard to Ewan, Top 10 is Grosu to Groenewegen, Top 15 is Cavendish to Appolonio).
The scenarios are the following :
S1: Only the sprint leaders. No other rider has more than 72 sprint. The startlist is the original one with those leadout sprinters removed and replaced by minor domestiques with decent flat stats. Question: Do the top sprinters generally have better results? And what kind of riders do the best on their own? (Maybe not that useful because this never happens, but I was interested in how this turns out and it might be a decent base line, and this is a hypothetical after all.)
S2: The Top 3 sprinters (Coquard, Swift, Degenkolb) each have their leadouts from the real startlist. Other than that, same start list as before (no leadouts). Q: Do the top sprinters have even better results with leadouts?
S3: The Top 3 sprinters have a more graduated sprint train to test Croatia's hypothesis of sprint train composition mattering. This includes one strong flat rider with modest SP/AC and one stronger but not high end sprinter. Q: What kind of leadout is better?
S4: Simply the real start list from the actual race this year (Preview here).
Original Startlist:
Spoiler
Szenario 1
No secondary sprinters whatsoever.
I'm going to limit this to a summary only because it was mostly chaotic sprints. The top sprinters were usually close to the front with one teammate leading them out, but they often went too early. Here are the summary statistics:
Scenario 1
Wins
Top 3
Top 5
Top 10
Outside Top 10
Top 5 Sprinters
1
12
6
13
18
Top 10 Sprinters
7
8
8
13
14
Top 15 Sprinters
2
1
12
Top 16 +
2
7
Conclusion:
The scenario was obviously not realistic in its composition, so don't interpret too much into the results. Breaks were caught pretty late, mostly with ~5kms to go and there were no real leadouts. The second tier of sprinters dominated the wins, most likely because the top tier was often kind of the leadout itself, but this is still a bad look, especially the number of times the top sprinters were outside the Top 10. Ahlstrand was hit the worst, with only three Top 10 appearances, Degenkolb and Ewan got 4 podiums each, Swift had the sole win.
Szenario 2
Festina, Aker and Iberia with sprint trains.
Round 1
8km
Podium Ambition and Volvo are the chasers. One escapee left up front, no trains yet.
5km
Escapees caught; still no trains.
3km
The sprint helpers we have aren't close to the front.
2km
PA and Volvo actually do form trains but they have no top sprinters other than Ewan and Ahlstrand.
1km
Ewan takes over under the red kite.
Finish line
Very close finish, but Ahlstrand takes it. Volvo train without real sprinter worked, Ewan did well too.
Round 1
1.
Ahlstrand
2.
Groenewegen
3.
Ewan
4.
Zabel
5.
Kennaugh
6.
Guardini
7.
Boeckmans
8.
Avelino
9.
Coquard
10.
Swift
Rounds 2 and 3:
Spoiler
Round 2
8km
Escapees are in sight but not caught yet. The Top 3 sprinters are bunched together without the leadout we expect them to have.
5km
Breakaway is caught, most sprinters somewhere behind the first line, no trains yet.
3km
Again it's PA and Volvo with the trains.
2km
Volvo takes the lead. Top 3 sprinters way behind.
1km
Novardianto brough Ewan to the front, he and Ahlstrand accelerate with 1km to go, Avelino currently in the lead.
Finish line
Ahlstrand with his second win, Ewan third again.
Round 2
1.
Ahlstrand
2.
Grosu
3.
Ewan
4.
Boeckmans
5.
Avelino
6.
Cavendish
7.
Guerao
8.
Kennaugh
9.
Nizzolo
10.
Oss
Round 3
8km
Breakaway was caught much earlier today. Top 3 sprinters are closer to the front, but still no leadout.
5km
Lots of sprinters somehwere from 10th position on. Grosu currently floating just behind the first line.
3km
The Top 3 made their way to the front and each have a teammate with them, but not one of the good sprinters they have ...
2km
Lammertink leads Grosu, but when the former starts sprinting, Grosu just ... doesn't.
1km
Grosu somehow found his speed again and now leads Groenewegen, Kennaugh and Vanderbiest. The PA rider is not Ewan and none of the two Volvo riders are Ahlstrand ...
Finish line
With a late charge, Kennaugh takes it.
Round 3
1.
Kennaugh
2.
Grosu
3.
Vanderbiest
4.
Coquard
5.
Groenewegen
6.
Zabel
7.
Swift
8.
Cavendish
9.
Guerao
10.
Ewan
Rounds 4-7:
Spoiler
Round 4
Round 5
Round 6
Round 7
Round 8
Round 9
Round 10
1.
Grosu
Groenewegen
Grosu
Ahlstrand
Grosu
Ewan
Ewan
2.
Avelino
Vanderbiest
Cavendish
Guardini
Groenewegen
Avelino
Kennaugh
3.
Cavendish
Grosu
Ewan
Groenewegen
Ahlstrand
Cavendish
Cavendish
4.
Zabel
Degenkolb
Kennaugh
Grosu
Ewan
Ahlstrand
Avelino
5.
Degenkolb
Guardini
Boeckmans
Swift
Avelino
Vanderbiest
Swift
6.
Kupfernagel
Cavendish
Avelino
Avelino
Cavendish
Grosu
Degenkolb
7.
Itami
Stauff
Groenewegen
Ewan
Boeckmans
Groenewegen
Boeckmans
8.
Guerao
Swift
Guerao
Guerao
Guerao
Zabel
Ahlstrand
9.
Afewerki
Boeckmans
Guardini
Cavendish
Kennaugh
Boeckmans
Vanderbiest
10.
Salleh
Avelino
Degenkolb
Coquard
Coquard
Guerao
Grosu
Statistics:
Scenario 2
Wins
Top 3
Top 5
Top 10
Outside Top 10
Top 5 Sprinters
5
4
7
11
23
Top 10 Sprinters
5
9
7
17
12
Top 15 Sprinters
6
3
15
Top 16 +
1
1
7
Conclusion:
Okay, on the one hand, this was an absolute desaster. Giving them "sprint trains" took Coquard, Swift and Degenkolb basically out of the race entirely, combining for five Top 5s and 17 (!!) places outside of the Top 10. I have no idea why this is and it seems game breaking, but remember that this is still an artificial startlist.
On the other hand, three teams, namely Volvo, Podium Ambitions and eBuddy, consistently managed to set up actualy sprint trains and did so successfully (Ahlstrand with three wins and a podium; Ewan with two wins, two podiums and only twice outside the Top 10; Grosu with three wins, three podiums and only once outside the Top 10). So what happens when the Top 3 get more modest sprint trains instead of just two good sprinters by their side?
Szenario 3
Festina, Aker and Iberia with more graduate sprint trains.
Instead of Stauff (80 SP) and Petit (80), Festina's two best sprint helpers are Weber (77) and Barbier (75).
Instead of Scully (79) and Haddi (77), Aker's best sprint helpers are Silva (77) and Oss (74).
Instead of Merino (78) and King (78), Iberia's best sprint helpers are Merino and Jans (74).
I had three rounds with screenshots again, but the results were basically the same and I decided to stop this scenario as it was clear that nothing had changed. Two good things I took from this:
a) Again, there were trains, and they were giving their leaders a chance to win (Ewan two podiums, Grosu won all three stages).
b) The teams setting them up were consistently the same (maybe a bit debatable whether this is a good thing, but at least there's not as much randomness I guess).
Adding on to that last point though, the big negatives were:
a) I could not tell why exactly those two to three teams (Volvo, PA, eBuddy) were the chosen by the game to be the ones leading the trains. I mean, the two final leadouts were among the better ones in the field, except for ...
b) ... the only "real" sprinters that were chosen as possible leadouts. They didn't do this once and their leaders seemed to be actively harmed by their presence.
I have to point out again that this was a bit of a weird startlist to begin with. So, to conclude, I took a few rounds with the actual startlist we had.
Szenario 4
Regular startlist.
Round 1
8km
Interestingly, now it's Aker and Iberia at the front of the peloton with a bit of Festina - as it should be.
5km
Break is caught. No trains formed yet, but many favorites have a sprinter teammate with them.
3km
Haddi and Scully for Swift, Bauhaus and Sbaragli for Groenewegen, Jans and King for Degenkolb a little further behind. Again, trains crossing, a little bit of distraction, but not too much.
2km
Sbaragli goes first. Coquard behind Stauff at the bottom of the screen.
1km
Scully jumps to the front but Swift has a gap to close. Grosu launches his sprint in the middle. Festina rider is Barbier.
Finish line
Groenewegen, Grosu and Swift a bit too early. Surprise win by Kennaugh ahead of Guardini.
Round 1
1.
Kennaugh
2.
Guardini
3.
Degenkolb
4.
Groenewegen
5.
Zabel
6.
Swift
7.
Grosu
8.
Vanderbiest
9.
Cavendish
10.
Barbier
Rounds 2 and 3:
Spoiler
Round 2
8km
Same teams up front, break already caught. Coquard and Swift nicely behind their sprinters, just need to get to the front in time.
5km
And they are coming closer to the front.
3km
This is what Festina and Swift want! Coquard had a small gap a few moments ago though, hopefully didn't have to waste too much energy.
2km
Trains crossing paths causes a major issue this time as Swift and Coquard get tangled up. Degenkolb behind King tries to exploit it.
1km
Weird stuff happening: Stauff came past everyone on the right, Coquard followed him. Everyone else went to Petits wheel, who simply slowed down.
Finish line
Consequently, dominating win for Festina. (Yes, this was a bit whacky, but the prep until ~1km to go was decent and showed that Aker and Festina can do something, and the results except for Stauff were not too wild, so I left it in this summary)
Round 2
1.
Coquard
2.
Stauff
3.
Grosu
4.
Avelino
5.
Degenkolb
6.
Groenewegen
7.
Silvestre
8.
Boeckmans
9.
Kip
10.
Guardini
Round 3
8km
It's Volvo and Podium Ambition at the helm of the pack today and indeed we see Ahlstrand and Ewan, who've been completely absent the first two rounds, towards the front.
5km
eBuddy is towards the front as well, are those three teams kind of linked? Ahlstrand in decent position already, Ewan with Nelson towards the bottom of the screen.
3km
Except for Ahlstrand in second position, this is all pretty messy. Grosu has a train on the left but Postlberger interrupts it. Novardianto and Alaphilippe going to the front for PA, but Nelson and Ewan still a bit boxed in. The Festina rider is Petit; Degenkolb, Swift and other nearby.
2km
eBuddy train on the left moving well now, Markus comes to the front on the right to support Ahlstrand. Ewan is the PA rider to the very bottom of the pic, just not getting ahead.
1km
Clarke led Kennaugh to the front, he accelerates now. Ahlstrand followed, also going, as does Grosu.
Finish line
In a close finish, Ahlstrand has the best legs. PA would have had a great shot had Ewan found the right wheel. Now they have Alaphilippe in fourth place. Just like in Round 1, Coquard disappears.
Round 3
1.
Ahlstrand
2.
Grosu
3.
Kennaugh
4.
Alaphilippe
5.
Swift
6.
Boeckmans
7.
Petit
8.
Degenkolb
9.
Silvestre
10.
Nelson
Rounds 4-10:
Spoiler
Round 4
Round 5
Round 6
Round 7
Round 8
Round 9
Round 10
1.
Ewan
Grosu
Ahlstrand
Ahlstrand
Ewan
Kennaugh
Groenewegen
2.
Grosu
Ewan
Kennaugh
Groenewegen
Ahlstrand
Boeckmans
Avelino
3.
Avelino
Boeckmans
Groenewegen
Avelino
Kennaugh
Ewan
Degenkolb
4.
Swift
Cavendish
Cavendish
Ewan
Avelino
Guardini
Cavendish
5.
Guardini
Ahlstrand
Ewan
Guardini
Grosu
Zabel
Kennaugh
6.
Cavendish
Degenkolb
Guardini
Boeckmans
Cavendish
Groenewegen
Ahlstrand
7.
Kennaugh
Coquard
Grosu
Guerao
Guardini
Cavendish
Vanderbiest
8.
Groenewegen
Swift
Degenkolb
Cavendish
Zabel
Grosu
Boeckmans
9.
Zabel
Zabel
Coquard
Grosu
Brus
Avelino
Guardini
10.
Boeckmans
Guerao
Boeckmans
Nelson
Coquard
Ahlstrand
Alaphilippe
For statistics, I assigned every rider points for their placements. 10 for the win, 9 for second and so on. I only did this for the Top 10 sprinters by sprint stat, so riders like Cavendish or Guardini don't appear here but they would be around 7th to 8th place I think. Here are the results:
1.
Grosu
55
Kennaugh
55
3.
Ahlstrand
52
4.
Ewan
50
5.
Groenewegen
47
6.
Avelino
41
7.
Degenkolb
33
8.
Swift
22
9.
Coquard
17
10.
Guerao
5
Now, again, this is obviously pretty bad when looking at Degenkolb, Swift and Coquard, especially considering 10 of Coquard's 17 points came from that weird win in round 2. After the first two rounds, Aker, Iberia and Festina again disappeared from the sprint preparations.
On the one hand, there's a decent randomness factor. No single rider has, according to these points, on average a better result than 5th. But I think that's the way it generally should be, because of so many sprinters and also because sprints are by nature the most volatile discipline, as has been mentioned here before.
On the other hand, there is still some consistency. eBuddy, Volvo and Podium Ambitions were almost always the ones with coordinated trains, and it shows. The trains worked generally well, rarely crashing into each other. Kennaugh and Groenewegen found consistently the right wheels directly behind one of those trains and profited.
The thing is that the game somehow decides who the teams in charge are, and the factors for that seem to be somehow fixed, because again I started a new stage race every time and within each scenario, it were the same teams for the most part - even in scenario 2, where they didn't even have leadouts. At the same time, in the first two rounds of the real startlist scenario, Iberia and Festina were the key teams, so there's a minimal chance for variance.
My impression is that if you have a team with a sprint train on paper, but you're not one of the three teams chosen to be in charge and form an actual sprint train, you're suffering. In the case of Coquard for instance, he was often around Petit and Stauff, but just somewhere around 40-50th position. Riders like Silvestre and Swift had similar issues at times.
In contrast, Groenewegen and Kennaugh also have at least one decent sprinter (although weaker - Bauhaus and Clarke, respectively, 78 sprint each). Groenewegen even had Sbaragli in addition, but they only very rarely formed trains, they just positioned themselves well close to the front and then followed the right riders. Other riders (Avelino, Cav, Guardini ...) were succesful that way, too, just less consistently.
I can test what happens when Swift or Coquard are the lone sprinters in their teams. I assume Volvo et al. should still run their trains, but maybe then Swift and Coquard would more likely be engaged in the sprint for themselves.
But that still would solve what I think is the key problem: I just can't figure out what makes certain teams be the dominant ones. Neither in the startlist nor the pre-stage preview, Ewan or Ahlstrand or Grosu were in the Top 3 favorites. Neither team has more dominant flat riders or sprinters than Festina or Iberia or Aker.
Getting to the bottom of this is in my opinion the main way to get some more predictability before a race about which sprinter should or could be successful and why someone like Coquard is so often underwhelming.
Do you have a rough number, in any/all scenarios, for the number of times the winner was the rider of the team that made a leadout vs number of times the winner was one of the sprinters tagging onto another train?
Thanks for a great effort in making some sense in all of this. It seems that it just doesn't make sense to be a 3 star favorite in this game anymore. Atleast those with young sprinters that could be 3 star favorite with some training no know to save the money as investing will only hurt them. Too bad for those of us who already spent 10+ mio in getting one.
In the second scenario, Ahlstrand, Ewan and Grosu combined for 8 of 10 stage wins. In the final scenario with the real start list, they combined for 7 of 10 stage wins. I didn't follow each stage after the first three in detail and mostly ran them through at x8, but Volvo, Podium Ambition and eBuddy were the main teams with trains on almost all stages. Kennaugh and Groenewegen were the only riders without leadouts who consistently finished well, but even then Groenewegen had leadouts at times.
It's basically the exact thing we have also seen this season. The top favorites either win or fail, while the riders you mention consistently perform well...
It seems very odd really.
It could be interesting to set all sprinters with the exact same stats for leadout riders... Just to see if it's still the same picture.
And one more question, were you running this as a brand new stage 1 each time, or as a single stage option like a classic? (can't quite tell from your intro)
I've done some tests too, but I think it will be better to post them tomorrow and focus on cunego's scenarios for now.
I also observed that Podium Ambition, Volvo and eBuddy were the team making trains most of the time. Others would too but not so often so I don't think it's random which teams are making the trains. I suspect it might have something to do with the strength of the entire team, but I don't have anything concrete to base this on other than PA being the strongest team in the race and almost always having a train.
TheManxMissile wrote:
And one more question, were you running this as a brand new stage 1 each time, or as a single stage option like a classic? (can't quite tell from your intro)
I was running this as a stage race, so all four stages selected, but only ever ran the first. And started a new stage race every time.
SotD wrote:
It could be interesting to set all sprinters with the exact same stats for leadout riders... Just to see if it's still the same picture.
Not sure I understand. Do you mean give every leadout rider on every team the same stats? Or give every sprint captain the same stats and see if the same teams still have the leadouts?
I plan to do one more set of maybe 5-10 stages with the only change being that I take Petit and Stauff from Coquard, as a case study if he then tries to get into the sprint on his own. I'm also going to do a few simulations with the default Cyanide database, but probably not tomorrow, this was a bit exhausting
Absolutely tremendous work cunego! Thank you so much for taking the time to run through these tests. That's a lot of work and is extremely helpful for all of us.
It seems this test has addressed two main questions:
1.) Sprint trains aren't hurting their sprinters, at least once we make the roads straight, wide and flat. That's good news!
2.) Whether you get to have a sprint train is not determined by actually having the strongest sprinter or team, according to Scenarios 2 and 3. That is not as good of news.
And we can also see that teams don't use the optimal combination of leadout riders for their trains, but that is nothing new for PCM. It has always been that way with AI as far as I can remember.
What I'm curious to figure out is why the game continually has the same teams setting up trains if its not based on team or individual strength. At least if setting up a train was fairly random among the top favorites each time, results might even out at least to some extent. But in this case it seems the game is choosing the teams by some invisible metric which favors the second tier riders. Figuring out what that metric would be key, but I have no idea how we would go about doing that, or if it's even possible.
Also interesting is that in Scenario 1, nobody made trains, but in Scenario 2+3, trains were made, but not by the stronger teams. If certain teams were simply predestined to make trains, why wouldn't they in Scenario 1 as well? Instead, they only did it when they were actually weaker in comparison. Perhaps there's some sort of stat variability which triggers train AI? I have no idea.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
cunego59 wrote:
I plan to do one more set of maybe 5-10 stages with the only change being that I take Petit and Stauff from Coquard, as a case study if he then tries to get into the sprint on his own.
This is something that I actually tested myself. Feel free to test it as well, but maybe you can go for other scenarios
I will send you a PM with what I've been testing in case you don't want to repeat the scenarios
TheManxMissile wrote:
And one more question, were you running this as a brand new stage 1 each time, or as a single stage option like a classic? (can't quite tell from your intro)
I was running this as a stage race, so all four stages selected, but only ever ran the first. And started a new stage race every time.
SotD wrote:
It could be interesting to set all sprinters with the exact same stats for leadout riders... Just to see if it's still the same picture.
Not sure I understand. Do you mean give every leadout rider on every team the same stats? Or give every sprint captain the same stats and see if the same teams still have the leadouts?
I plan to do one more set of maybe 5-10 stages with the only change being that I take Petit and Stauff from Coquard, as a case study if he then tries to get into the sprint on his own. I'm also going to do a few simulations with the default Cyanide database, but probably not tomorrow, this was a bit exhausting
All leadout riders the same stats. So we isolate the sprinters with the exact same setup around them... If it's still the same sprinters getting the train then we'll know that 3 star favorites will not make trains, however the 2 star once will.
I'm really happy that you do this amount of work mate.
SotD wrote:
All leadout riders the same stats. So we isolate the sprinters with the exact same setup around them... If it's still the same sprinters getting the train then we'll know that 3 star favorites will not make trains, however the 2 star once will.
I'm really happy that you do this amount of work mate.
That's a good idea. While I'm thinking about it, it could also be interesting to swap the stats of for instance Coquard, Degenkolb and Swift with Grosu, Ahlstrand and Ewan to see if the cause for making sprint trains is actually being a 2 star favorite or if it has more to do with team composition.
That being said, for scenarios with stat changes, it would be cool if someone else could edit the DB accordingly. That would take a little bit of the work off my hands and I could focus on just running the races. Don't know if the ManGame cdb is available to anyone, but maybe a fellow reporter can do the edits
I also want to echo bbl: It really does seem that sprint trains work, which is really neat. We just have to figure who sets them and why.
True. Just swapping the lead sprinter for the top 3 favorites and the 3 most succesful sprinters would do the trick. If they still get the train it's all about the sprinter, while if the 3 star favorites get the train it's about the team. That would be very interesting data.
quadsas wrote:
What happens if you try to survice and form a train with control team?
Also difficulty. Were not quick simming and theres a human team so difficulty affects cpu riders
This is a good point. We've always kept it at normal in the past, but we never (at least to my knowledge) investigated how difficulty might affect AI differently in this version.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
quadsas wrote:
What happens if you try to survice and form a train with control team?
Also difficulty. Were not quick simming and theres a human team so difficulty affects cpu riders
This is a good point. We've always kept it at normal in the past, but we never (at least to my knowledge) investigated how difficulty might affect AI differently in this version.
Difficulty is still part of the .cdb and still does the same thing, just reducing users rider attributes based on level.
quadsas wrote:
What happens if you try to survice and form a train with control team?
The control team is far too weak to take part in the sprint. Also, its riders are usually retired at the beginning of the stage to so that they don't effect the race.
The difficulty was on normal, as is standard. Like Kent says, as far as I know it only changes the human attributes and doesn't interfere with the AI, and again, the human team is taken out of the equation as much as possible. Not sure what influence that would have, and even if there was something, what could be done about it.
I've been running some tests using the flat stage provided by Tamijo and the startlist from the Barbados Cycling Festival. The dataset is too small to make any certain conclusions, but I do believe it gives a better understanding on what works and what does not.
After some tests, I've started doing changes to the start list that will be explained as they were made. The changes focused on Podium Ambition (which I believed to have had the strongest setup in the race altogether) and Festina, which arguabally have the best sprinter in the game.
Every test also has some notes of things I observed during the sprint. Sorry for not redacting them and the typos in this post as I was quite tired after having run this thing 20 times. They should still give you a good idea of what was going on.
test1: stauff and petit tire themselves out very early and leave coquard alone, scully takes over who also tires himself out, then other two aker riders
Spoiler
1
John Degenkolb
Iberia - Team Degenkolb
3h55'14
2
Bryan Coquard
Festina - OAKA
s.t.
3
Ben Swift
Aker - MOT
s.t.
4
Rick Zabel
Desigual
s.t.
5
Caleb Ewan
Podium Ambition
s.t.
6
Eduard Grosu
eBuddy
s.t.
7
Romain Vanderbiest
Carrefour - ESPN
s.t.
8
Oscar Avelino
Duolingo
s.t.
9
Peter Kennaugh
Air France - KLM
s.t.
10
Kris Boeckmans
Campari/Asahi development
s.t.
11
Dylan Groenewegen
Aegon - Lavazza
s.t.
12
Oscar Guerao
SPAR - Siam Cement
s.t.
13
Andrea Guardini
Fablok - Bank BGZ
s.t.
14
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
15
Fabio Silvestre
Berg Cycles
s.t.
16
Janis Dakteris
Aker - MOT
s.t.
17
Mark Cavendish
Kraftwerk Man Machine
s.t.
18
Pedro Merino Criado
Iberia - Team Degenkolb
s.t.
19
Maurits Lammertink
eBuddy
s.t.
test2: podium ambition set up quite an impressive train for ewan. Most sprinters (grosu, degenkolb, swift) followed it with. Coquard chose to follow his own train, but they could not really breakaway from the pack and make an impact and they stayed behind those following the Podium Ambition train.
Inside the last kilometer Roeledants had a fantastic outburst of speed to leave the train Podium ambition train way behind as he was leading out for Ahlstrand. He even looked like winning but he was completely empty and slowed down quit a bit right before the finish line.
Spoiler
1
Jonas Ahlstrand
Volvo acc. by Spotify
3h59'16
2
Romain Vanderbiest
Carrefour - ESPN
s.t.
3
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
4
John Degenkolb
Iberia - Team Degenkolb
s.t.
5
Caleb Ewan
Podium Ambition
s.t.
6
Fabio Silvestre
Berg Cycles
s.t.
7
Ben Swift
Aker - MOT
s.t.
8
Eduard Grosu
eBuddy
s.t.
9
Mark Cavendish
Kraftwerk Man Machine
s.t.
10
Kenji Itami
Rakuten Pro Cycling
s.t.
11
Bryan Coquard
Festina - OAKA
s.t.
12
Oscar Avelino
Duolingo
s.t.
13
Peter Kennaugh
Air France - KLM
s.t.
14
Kris Boeckmans
Campari/Asahi development
s.t.
15
Oscar Guerao
SPAR - Siam Cement
s.t.
16
Danilo Kupfernagel
Berg Cycles
s.t.
17
Andrea Guardini
Fablok - Bank BGZ
s.t.
18
Maurits Lammertink
eBuddy
s.t.
19
Willie Smit
EA Vesuvio
s.t.
20
Dylan Groenewegen
Aegon - Lavazza
s.t.
test3: oss for swift vs festina. Everyone seems to be on the festina train, except ewan and grosu who trusted their leadouts but they are way behind the festina train was amazingly fast. Even swift jumped wheels, but they left cuquard out alone with 1km to go and it was clearly too soon.
Spoiler
1
Peter Kennaugh
Air France - KLM
4h02'42
2
Dylan Groenewegen
Aegon - Lavazza
s.t.
3
Mark Cavendish
Kraftwerk Man Machine
s.t.
4
John Degenkolb
Iberia - Team Degenkolb
s.t.
5
Oscar Avelino
Duolingo
s.t.
6
Bryan Coquard
Festina - OAKA
s.t.
7
Ben Swift
Aker - MOT
s.t.
8
Kenji Itami
Rakuten Pro Cycling
s.t.
9
Oscar Guerao
SPAR - Siam Cement
s.t.
10
Romain Vanderbiest
Carrefour - ESPN
s.t.
11
Eduard Grosu
eBuddy
s.t.
12
Kris Boeckmans
Campari/Asahi development
s.t.
13
Andreas Stauff
Festina - OAKA
s.t.
14
Tom Scully
Aker - MOT
s.t.
15
Marco Brus
eBuddy
s.t.
16
Ricki Nelson
Podium Ambition
s.t.
17
Julian Alaphilippe
Podium Ambition
s.t.
18
Rick Zabel
Desigual
s.t.
19
Caleb Ewan
Podium Ambition
s.t.
20
Giacomo Nizzolo
Repsol - Netflix
s.t.
test 4: no trains, iberia tried something but failed with people attacking in the last kilometers. 2km to go Nelson puts on an amazing sprint adn Ewan follows him. Those that were quick to jump on than train (Grosu, Swift and Zabel) all tire themselves out as Ewan is left alone 1k to go
Spoiler
1
Jonas Ahlstrand
Volvo acc. by Spotify
3h57'03
2
John Degenkolb
Iberia - Team Degenkolb
s.t.
3
Fabio Silvestre
Berg Cycles
s.t.
4
Dylan Groenewegen
Aegon - Lavazza
s.t.
5
Ben Swift
Aker - MOT
s.t.
6
Mark Cavendish
Kraftwerk Man Machine
s.t.
7
Rick Zabel
Desigual
s.t.
8
Oscar Avelino
Duolingo
s.t.
9
Andrea Guardini
Fablok - Bank BGZ
s.t.
10
Caleb Ewan
Podium Ambition
s.t.
11
Peter Kennaugh
Air France - KLM
s.t.
12
Ismael Kip
Andorra Cycling Project
s.t.
13
Oscar Guerao
SPAR - Siam Cement
s.t.
14
Eduard Grosu
eBuddy
s.t.
15
Danilo Kupfernagel
Berg Cycles
s.t.
16
Yudai Arashiro
Rakuten Pro Cycling
s.t.
17
Davide Appollonio
Carlsberg - Danske Bank
s.t.
18
Maxime Vantomme
Minions
s.t.
19
Benedikt Mundle
Sauber Petronas Racing
s.t.
20
Patrick Naud
Campari/Asahi development
s.t.
test 5: 3 clear trains with 3 riders by Festina, Aker and Iberia. Podium are also trying but they are not as fast
Spoiler
1
John Degenkolb
Iberia - Team Degenkolb
4h01'58
2
Dylan Groenewegen
Aegon - Lavazza
s.t.
3
Bryan Coquard
Festina - OAKA
s.t.
4
Ben Swift
Aker - MOT
s.t.
5
Kris Boeckmans
Campari/Asahi development
s.t.
6
Ricki Nelson
Podium Ambition
s.t.
7
Caleb Ewan
Podium Ambition
s.t.
8
Oscar Guerao
SPAR - Siam Cement
s.t.
9
Peter Kennaugh
Air France - KLM
s.t.
10
Mark Cavendish
Kraftwerk Man Machine
s.t.
11
Oscar Avelino
Duolingo
s.t.
12
Eduard Grosu
eBuddy
s.t.
13
Andrea Guardini
Fablok - Bank BGZ
s.t.
14
Yudai Arashiro
Rakuten Pro Cycling
s.t.
15
Pedro Merino Criado
Iberia - Team Degenkolb
s.t.
16
Andreas Stauff
Festina - OAKA
s.t.
17
Tom Scully
Aker - MOT
s.t.
18
Kenji Itami
Rakuten Pro Cycling
s.t.
19
Maurits Lammertink
eBuddy
s.t.
20
Ho-Ting Kwok
Air France - KLM
s.t.
Starlist change From my experience with the game, it might be better if the first rider in the 3men train is a good flat rider, not necesarely a sprinter. Floris Gerts and Fernando Gaviria replace Petit and Stauff in the Festina line-up
test 6: podium ambition set up a very high pace in the pack and made a split. They seem to be the only one with leadouts left in the group. Coquard is asking for water with 7 km to go. Podium control the sprint with everyone following them.
Nelson was forced to leave Ewan alone quite early. Coquard was really fast inside the last kilometer and could have won but there was no gap for him to get trhough in a very close finish
Spoiler
1
Rick Zabel
Desigual
3h54'58
2
Mark Cavendish
Kraftwerk Man Machine
s.t.
3
Romain Vanderbiest
Carrefour - ESPN
s.t.
4
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
5
Oscar Avelino
Duolingo
s.t.
6
Bryan Coquard
Festina - OAKA
s.t.
7
Caleb Ewan
Podium Ambition
s.t.
8
Oscar Guerao
SPAR - Siam Cement
s.t.
9
John Degenkolb
Iberia - Team Degenkolb
s.t.
10
Ben Swift
Aker - MOT
s.t.
11
Andrea Guardini
Fablok - Bank BGZ
s.t.
12
Eduard Grosu
eBuddy
s.t.
13
Peter Kennaugh
Air France - KLM
s.t.
14
Kris Boeckmans
Campari/Asahi development
s.t.
15
Kenji Itami
Rakuten Pro Cycling
s.t.
16
Fabio Silvestre
Berg Cycles
s.t.
17
Yudai Arashiro
Rakuten Pro Cycling
s.t.
18
Maurits Lammertink
eBuddy
s.t.
19
Juan Abenhamar Gallego Martin
Iberia - Team Degenkolb
s.t.
20
Benedikt Mundle
Sauber Petronas Racing
s.t.
test 7: no coherent trains. eBuddy, podium, iberia and volvo try to form but nothing conherent and it's pretty much everyone for himself inside the last kilometer.
Spoiler
1
Ben Swift
Aker - MOT
3h58'04
2
Peter Kennaugh
Air France - KLM
s.t.
3
Eduard Grosu
eBuddy
s.t.
4
Bryan Coquard
Festina - OAKA
s.t.
5
Romain Vanderbiest
Carrefour - ESPN
s.t.
6
Mark Cavendish
Kraftwerk Man Machine
s.t.
7
John Degenkolb
Iberia - Team Degenkolb
s.t.
8
Danilo Kupfernagel
Berg Cycles
s.t.
9
Fabio Silvestre
Berg Cycles
s.t.
10
Oscar Guerao
SPAR - Siam Cement
s.t.
11
Kenji Itami
Rakuten Pro Cycling
s.t.
12
Caleb Ewan
Podium Ambition
s.t.
13
Dylan Groenewegen
Aegon - Lavazza
s.t.
14
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
15
Rick Zabel
Desigual
s.t.
16
Kris Boeckmans
Campari/Asahi development
s.t.
17
Mohamed Harrif Salleh
Sauber Petronas Racing
s.t.
18
Ho-Ting Kwok
Air France - KLM
s.t.
19
Pedro Merino Criado
Iberia - Team Degenkolb
s.t.
20
Marco Brus
eBuddy
s.t.
Start list change Gerts is not doing any attempts at helping with the Festina train. Time to replace him with a better sprinter. Van Avermaet in, Gerts out
test 8: eBuddy, volvo, podium trains
Spoiler
1
John Degenkolb
Iberia - Team Degenkolb
3h59'48
2
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
3
Oscar Guerao
SPAR - Siam Cement
s.t.
4
Oscar Avelino
Duolingo
s.t.
5
Bryan Coquard
Festina - OAKA
s.t.
6
Caleb Ewan
Podium Ambition
s.t.
7
Ho-Ting Kwok
Air France - KLM
s.t.
8
Peter Kennaugh
Air France - KLM
s.t.
9
Eduard Grosu
eBuddy
s.t.
10
Yudai Arashiro
Rakuten Pro Cycling
s.t.
11
Benedikt Mundle
Sauber Petronas Racing
s.t.
12
Andrea Guardini
Fablok - Bank BGZ
s.t.
13
Kris Boeckmans
Campari/Asahi development
s.t.
14
Mohamed Harrif Salleh
Sauber Petronas Racing
s.t.
15
Ricki Nelson
Podium Ambition
s.t.
16
Kenji Itami
Rakuten Pro Cycling
s.t.
17
Ismael Kip
Andorra Cycling Project
s.t.
18
Marco Brus
eBuddy
s.t.
19
Vincente Garcia De Mateos
Control Team
s.t.
20
Romain Vanderbiest
Carrefour - ESPN
s.t.
Startlist change Podium Ambition seem to always be at the front and more often than not they are setting up a sprint train. It looks to me like Nelson's low resistence stat is making him only stay in front for a very short time, leaving Ewan alone way too early. Nelson out, Lahcen Saber in
test 9: everyone seems to be all over the place. Trains are trying to be formed but no one can get their neck out. Gaviria is finally with Coquard. Swift is the only one to seem to be getting a proper lead out but is left out in the wind too early
Spoiler
1
Caleb Ewan
Podium Ambition
3h57'21
2
Eduard Grosu
eBuddy
s.t.
3
Oscar Guerao
SPAR - Siam Cement
s.t.
4
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
5
Mark Cavendish
Kraftwerk Man Machine
s.t.
6
Kris Boeckmans
Campari/Asahi development
s.t.
7
Fabio Silvestre
Berg Cycles
s.t.
8
Ben Swift
Aker - MOT
s.t.
9
John Degenkolb
Iberia - Team Degenkolb
s.t.
10
Andrea Guardini
Fablok - Bank BGZ
s.t.
11
Fernando Gaviria
Festina - OAKA
s.t.
12
Bryan Coquard
Festina - OAKA
s.t.
13
Romain Vanderbiest
Carrefour - ESPN
s.t.
14
Peter Kennaugh
Air France - KLM
s.t.
15
Oscar Avelino
Duolingo
s.t.
16
Lahcen Saber
Podium Ambition
s.t.
17
Dylan Groenewegen
Aegon - Lavazza
s.t.
18
Danilo Kupfernagel
Berg Cycles
s.t.
19
Kenji Itami
Rakuten Pro Cycling
s.t.
20
Ho-Ting Kwok
Air France - KLM
s.t.
test 10: volve, ebuddy, podium. Coquard behind ewan, swift behind grosu
Spoiler
1
Ben Swift
Aker - MOT
3h59'56
2
Bryan Coquard
Festina - OAKA
s.t.
3
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
4
Caleb Ewan
Podium Ambition
s.t.
5
Eduard Grosu
eBuddy
s.t.
6
Oscar Avelino
Duolingo
s.t.
7
Dylan Groenewegen
Aegon - Lavazza
s.t.
8
Mohamed Harrif Salleh
Sauber Petronas Racing
s.t.
9
Oscar Guerao
SPAR - Siam Cement
s.t.
10
Mark Cavendish
Kraftwerk Man Machine
s.t.
11
Kenji Itami
Rakuten Pro Cycling
s.t.
12
John Degenkolb
Iberia - Team Degenkolb
s.t.
13
Lahcen Saber
Podium Ambition
s.t.
14
Ho-Ting Kwok
Air France - KLM
s.t.
15
Marco Brus
eBuddy
s.t.
16
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
17
Kris Boeckmans
Campari/Asahi development
s.t.
18
Rick Zabel
Desigual
s.t.
19
Peter Kennaugh
Air France - KLM
s.t.
20
Romain Vanderbiest
Carrefour - ESPN
s.t.
Start list change Coquard is rarely getting a train, while Ewan almost always has one. Is it the team or the sprinter responsible for this? Ewan and Coquard switch teams
test 11: festina do make a train this time (barbier, gaviria, ewan). they and podium are pretty much the strongest
Spoiler
1
Bryan Coquard
Podium Ambition
3h55'10
2
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
3
Peter Kennaugh
Air France - KLM
s.t.
4
Caleb Ewan
Festina - OAKA
s.t.
5
Dylan Groenewegen
Aegon - Lavazza
s.t.
6
Mark Cavendish
Kraftwerk Man Machine
s.t.
7
Oscar Guerao
SPAR - Siam Cement
s.t.
8
Fabio Silvestre
Berg Cycles
s.t.
9
Romain Vanderbiest
Carrefour - ESPN
s.t.
10
Ho-Ting Kwok
Air France - KLM
s.t.
11
Lahcen Saber
Podium Ambition
s.t.
12
Fernando Gaviria
Festina - OAKA
s.t.
13
Eduard Grosu
eBuddy
s.t.
14
Kenji Itami
Rakuten Pro Cycling
s.t.
15
Davide Appollonio
Carlsberg - Danske Bank
s.t.
16
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
17
Maurits Lammertink
eBuddy
s.t.
18
Kris Boeckmans
Campari/Asahi development
s.t.
19
Ben Swift
Aker - MOT
s.t.
20
Andrea Guardini
Fablok - Bank BGZ
s.t.
test 12: volvo, ebuddy and podium. ewan isolated and quite far back
Spoiler
1
Bryan Coquard
Podium Ambition
3h58'16
2
Ben Swift
Aker - MOT
s.t.
3
Eduard Grosu
eBuddy
s.t.
4
Dylan Groenewegen
Aegon - Lavazza
s.t.
5
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
6
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
7
Mark Cavendish
Kraftwerk Man Machine
s.t.
8
John Degenkolb
Iberia - Team Degenkolb
s.t.
9
Oscar Avelino
Duolingo
s.t.
10
Kris Boeckmans
Campari/Asahi development
s.t.
11
Maxime Vantomme
Minions
s.t.
12
Fabio Silvestre
Berg Cycles
s.t.
13
Ho-Ting Kwok
Air France - KLM
s.t.
14
Lahcen Saber
Podium Ambition
s.t.
15
Maurits Lammertink
eBuddy
s.t.
16
Sten Stenersen
Volvo acc. by Spotify
s.t.
17
Dennis Van Winden
SPAR - Siam Cement
s.t.
18
Patrick Naud
Campari/Asahi development
s.t.
19
Juan Abenhamar Gallego Martin
Iberia - Team Degenkolb
s.t.
20
Johann van Zyl
Kraftwerk Man Machine
s.t.
test 13: aker, ebuddy and volvo are trying but everything is too packed toghether and everyone is all over the place
Spoiler
1
Ben Swift
Aker - MOT
3h57'02
2
Eduard Grosu
eBuddy
s.t.
3
Fabio Silvestre
Berg Cycles
s.t.
4
Rick Zabel
Desigual
s.t.
5
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
6
Yudai Arashiro
Rakuten Pro Cycling
s.t.
7
Marildo Yzeiraj
Aker - MOT
s.t.
8
Oscar Avelino
Duolingo
s.t.
9
Sten Stenersen
Volvo acc. by Spotify
s.t.
10
Dennis Van Winden
SPAR - Siam Cement
s.t.
11
Benedikt Mundle
Sauber Petronas Racing
s.t.
12
Johann van Zyl
Kraftwerk Man Machine
s.t.
13
Peter Kennaugh
Air France - KLM
s.t.
14
Patrick Naud
Campari/Asahi development
s.t.
15
Oscar Guerao
SPAR - Siam Cement
s.t.
16
Danilo Kupfernagel
Berg Cycles
s.t.
17
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
18
Laurens De Vreese
Kraftwerk Man Machine
s.t.
19
Maurits Lammertink
eBuddy
s.t.
20
Elias Afewerki
EA Vesuvio
s.t.
test 14: ebuddy and podium. other sprinters happy to follow either train
Spoiler
1
Ben Swift
Aker - MOT
4h05'43
2
Bryan Coquard
Podium Ambition
s.t.
3
Eduard Grosu
eBuddy
s.t.
4
Dylan Groenewegen
Aegon - Lavazza
s.t.
5
Peter Kennaugh
Air France - KLM
s.t.
6
Mark Cavendish
Kraftwerk Man Machine
s.t.
7
John Degenkolb
Iberia - Team Degenkolb
s.t.
8
Kris Boeckmans
Campari/Asahi development
s.t.
9
Romain Vanderbiest
Carrefour - ESPN
s.t.
10
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
11
Caleb Ewan
Festina - OAKA
s.t.
12
Danilo Kupfernagel
Berg Cycles
s.t.
13
Rick Zabel
Desigual
s.t.
14
Andrea Guardini
Fablok - Bank BGZ
s.t.
15
Yudai Arashiro
Rakuten Pro Cycling
s.t.
16
Oscar Avelino
Duolingo
s.t.
17
Oscar Guerao
SPAR - Siam Cement
s.t.
18
Ho-Ting Kwok
Air France - KLM
s.t.
19
Fabio Silvestre
Berg Cycles
s.t.
20
Ismael Kip
Andorra Cycling Project
s.t.
test 15: only one train and everyone is following it: volvo. Coquard quite behind in the pecking order. Festina tried to put something together but it failed apart and Ewan is further back
Spoiler
1
John Degenkolb
Iberia - Team Degenkolb
3h56'59
2
Ben Swift
Aker - MOT
s.t.
3
Dylan Groenewegen
Aegon - Lavazza
s.t.
4
Eduard Grosu
eBuddy
s.t.
5
Bryan Coquard
Podium Ambition
s.t.
6
Kenji Itami
Rakuten Pro Cycling
s.t.
7
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
8
Peter Kennaugh
Air France - KLM
s.t.
9
Rick Zabel
Desigual
s.t.
10
Oscar Guerao
SPAR - Siam Cement
s.t.
11
Romain Vanderbiest
Carrefour - ESPN
s.t.
12
Mark Cavendish
Kraftwerk Man Machine
s.t.
13
Kris Boeckmans
Campari/Asahi development
s.t.
14
Oscar Avelino
Duolingo
s.t.
15
Caleb Ewan
Festina - OAKA
s.t.
16
Andrea Guardini
Fablok - Bank BGZ
s.t.
17
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
18
Lahcen Saber
Podium Ambition
+ 45
19
Fabio Silvestre
Berg Cycles
s.t.
20
Willie Smit
EA Vesuvio
s.t.
Start list change Coquard seems to be more consistent since joining Podium Ambition and very strong when Saber has been lead in him out. Saber is a beast for sure, but is he the sole man responsible for this? Time to replace him with a weaker lead out. Saber out, Jon Aberasturi in
test 16: podium vs ebuddy. swift is behind podium, groenewegen behind ebuddy. Ewan caught further back
Spoiler
1
Bryan Coquard
Podium Ambition
3h56'51
2
Peter Kennaugh
Air France - KLM
s.t.
3
John Degenkolb
Iberia - Team Degenkolb
s.t.
4
Eduard Grosu
eBuddy
s.t.
5
Dylan Groenewegen
Aegon - Lavazza
s.t.
6
Ben Swift
Aker - MOT
s.t.
7
Kris Boeckmans
Campari/Asahi development
s.t.
8
Oscar Avelino
Duolingo
s.t.
9
Rick Zabel
Desigual
s.t.
10
Maxime Vantomme
Minions
s.t.
11
Oscar Guerao
SPAR - Siam Cement
s.t.
12
Ho-Ting Kwok
Air France - KLM
s.t.
13
Romain Vanderbiest
Carrefour - ESPN
s.t.
14
Caleb Ewan
Festina - OAKA
s.t.
15
Andrea Guardini
Fablok - Bank BGZ
s.t.
16
Davide Appollonio
Carlsberg - Danske Bank
s.t.
17
Maurits Lammertink
eBuddy
s.t.
18
Jon Aberasturi
Podium Ambition
s.t.
19
Patrick Naud
Campari/Asahi development
s.t.
20
Fabio Silvestre
Berg Cycles
s.t.
test 17: volvo dominated this by a long shot. Grosu right behind them, Coquard is brought in Grosu's wheel by a team - mate. Others follow in line
Spoiler
1
Eduard Grosu
eBuddy
4h02'44
2
John Degenkolb
Iberia - Team Degenkolb
s.t.
3
Caleb Ewan
Festina - OAKA
s.t.
4
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
5
Ben Swift
Aker - MOT
s.t.
6
Bryan Coquard
Podium Ambition
s.t.
7
Andrea Guardini
Fablok - Bank BGZ
s.t.
8
Mark Cavendish
Kraftwerk Man Machine
s.t.
9
Dylan Groenewegen
Aegon - Lavazza
s.t.
10
Rick Zabel
Desigual
s.t.
11
Mohamed Harrif Salleh
Sauber Petronas Racing
s.t.
12
Oscar Avelino
Duolingo
s.t.
13
Romain Vanderbiest
Carrefour - ESPN
s.t.
14
Oscar Guerao
SPAR - Siam Cement
s.t.
15
Kris Boeckmans
Campari/Asahi development
s.t.
16
Ho-Ting Kwok
Air France - KLM
s.t.
17
Kenji Itami
Rakuten Pro Cycling
s.t.
18
Danilo Kupfernagel
Berg Cycles
s.t.
19
Jurgen Roelandts
Volvo acc. by Spotify
s.t.
20
Marildo Yzeiraj
Aker - MOT
s.t.
Startlist change Festina seem to be doing even worse in terms of train with this set up than their original lead-out. Time to make it back to 3 sprinters. Petit in, Van Avermaet out
test 18: festina vs podium. grosu and groenewegen behind podium. degenkolb behind podium
Spoiler
1
Bryan Coquard
Podium Ambition
3h58'22
2
John Degenkolb
Iberia - Team Degenkolb
s.t.
3
Caleb Ewan
Festina - OAKA
s.t.
4
Eduard Grosu
eBuddy
s.t.
5
Oscar Guerao
SPAR - Siam Cement
s.t.
6
Ben Swift
Aker - MOT
s.t.
7
Dylan Groenewegen
Aegon - Lavazza
s.t.
8
Rick Zabel
Desigual
s.t.
9
Jonas Ahlstrand
Volvo acc. by Spotify
s.t.
10
Mark Cavendish
Kraftwerk Man Machine
s.t.
11
Oscar Avelino
Duolingo
s.t.
12
Andrea Guardini
Fablok - Bank BGZ
s.t.
13
Romain Vanderbiest
Carrefour - ESPN
s.t.
14
Coen Vermeltfoort
eBuddy
s.t.
15
Fabio Silvestre
Berg Cycles
s.t.
16
Fernando Gaviria
Festina - OAKA
s.t.
17
Peter Kennaugh
Air France - KLM
s.t.
18
Kenji Itami
Rakuten Pro Cycling
s.t.
19
Maurits Lammertink
eBuddy
s.t.
20
Kris Boeckmans
Campari/Asahi development
s.t.
test 19: 4 trains: ebuddy and podium slightly better than festina and aker
Spoiler
1
Bryan Coquard
Podium Ambition
4h12'56
2
Eduard Grosu
eBuddy
s.t.
3
Ben Swift
Aker - MOT
s.t.
4
Mark Cavendish
Kraftwerk Man Machine
s.t.
5
Caleb Ewan
Festina - OAKA
s.t.
6
Tom Scully
Aker - MOT
s.t.
7
Mohamed Harrif Salleh
Sauber Petronas Racing
s.t.
8
Kenji Itami
Rakuten Pro Cycling
s.t.
9
Romain Vanderbiest
Carrefour - ESPN
s.t.
10
Andrea Guardini
Fablok - Bank BGZ
s.t.
11
Peter Kennaugh
Air France - KLM
s.t.
12
Fernando Gaviria
Festina - OAKA
s.t.
13
Dylan Groenewegen
Aegon - Lavazza
s.t.
14
Oscar Guerao
SPAR - Siam Cement
s.t.
15
Danilo Kupfernagel
Berg Cycles
s.t.
16
Kris Boeckmans
Campari/Asahi development
s.t.
17
Jon Aberasturi
Podium Ambition
s.t.
18
Oscar Avelino
Duolingo
s.t.
19
Rick Zabel
Desigual
s.t.
20
Marco Brus
eBuddy
s.t.
test 20: festina, ebuddy and volvo, with podium putting up a great fight to make up the ground
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
