It seems equally random to me... Now I have only glimpsed, and primarily is looking for Coquard. Obviously I like Test 1 the best, but still he have that 15th place aswell as his win which makes it seem more random than anything... To see him have the same level as riders like Boeckmans, Kennaugh, Vanderbiest, Silvestre, Groenewegen, Itami and Guardini is not good really.
Obviously he can lose to riders like Swift, Ahlstrand, Cavendish, Degenkolb and Grosu (those kind of riders), but those are also all over the place.
Ahlstrand seems to be the most consistent rider - which is OK I guess, as his stats are generally good, but it's difficult to see what the statistically stat math really is other than randomness.
Croatia14 wrote:
Tests looking really good and how they should be. We're the trains set diverse, meaning different riders had trains?
Do you really think so? The 3 best sprinters in the bunch shared a total amount of 6 stagewins between them from 15 tries. That is aproximately 1/3 of the time that the biggest stars win. That is indebately too little in my oppinion!
These riders:
Caleb Ewan
Peter Kennaugh
Jonas Ahlstrand
Mark Cavendish
Fabio Silvestre
Kris Boeckmans
Performed equally strong or better. I know we can't have the best sprinter win all the time, but we most certainly didn't have so in PCM 15, and to see them only fight for wins every 1/3 of the time is simply to little, when it's split between the 3 best.
Succesrates for top 3 sprinters must be top 3, while everything outside top 5 is a failure.
John Degenkolb (Succesratio: 27%) (Failure: 67%) (Outside top 10: 33%)
7-8-10-2-11-1-UNR-5-UNR-1-13-1-8-7-8
Ben Swift (Succesratio: 20%) (Failure: 67%) (Outside top 10: 40%)
11-UNR-4-UNR-13-7-4-7-8-UNR-10-3-12-2-1
So in 2/3rd of the time the topriders doesn't just fail to meet their succescriteria - they also heavily fail as they are not in top 5 - with also a huge margin of results outside the top 10. So having such as rider with the leadoutrains etc. you can expect to succed in your goals aproximately 1/5 of the time, which simply is too little. If Taaramae would miss top 3 in 4/5 races he attended we wouldn't even discuss whether or not this game would be cut. That would simply be game-breaking behaviour. You have to see this as Taaramae more often than not fucking up and ending outside the top 10 behind riders with 81-82MO.
These are the results of the riders rated 4-8th. Their succescriteria must be top 5 (In reality should probably be top 8), while failure is outside the top 10.
Jonas Ahlstrand (Succesratio: 47%) (Failure: 33%)
13-7-11-6-4-5-11-1-3-5-1-12-6-1-11
Caleb Ewan (Succesratio: 30%) (Failure: 20%)
15-1-9-1-1-11-9-UNR-UNR-UNR-UNR-2-9-7 * I believe he wasn't in the startlist on test 3, so those results are cut.
Eduard Grosu (Succesratio: 47%) (Failure: 13%)
8-4-6-5-8-2-6-13-4-7-3-10-13-3-5
Mark Cavendish (Succesratio: 40%) (Failure: 40%)
1-12-3-4-9-9-3-8-13-UNR-UNR-UNR-5-15-3 * Could be some crash thing that hasn't been shut off for test 3? They are included though...
Test looking really bad, and shows exactly what we have already seen. It is SIGNIFICANTLY better to have the sprinters who are rated 4-8th in a race than to have those who are rated 1-3. Not only do their have a higher succesratio, they also have a significantly lower fail-ratio, so they just perform so much more in terms of continuity.
Like said, we wouldn't accept to see Taaramae, Pluchkin and Herklotz were all over the place, and could end up 10th in a GT while riders like Lecuisinier, Denifl, Gesink and Sicard were to sit way more often on the top spots.
Edited by SotD on 24-04-2020 08:42
Yes I do. I don't think it's bad that Ewan, Silvestre, Kennaugh and Ahlstrand perform equal/better, in fact the high flat sprinters were just massively underrated in PCM15. Also it's down to the fact that the top sprinters just don't have good sprint trains right now, because they didn't need them. They don't perform as dominant as in PCM15, but I think that's very good.
I think the composition of a good sprint train is the key to success for those teams, especially one that doesn't wear their main sprinters out. Teams with Top sprinters now need to emphasis their team more on those, which is a good thing imo.
I do think that is mostly right... teams are expected to work as teams in 18 and on... Which means for teams with favorites having some riders that specialize in controlling the race, either good climbers, or in many cases better flat specialists to help bring back the breakaways.
Ok so first look through: ugh that is awful!
The variation between stages is all over the place for every rider! (i will offer some genuine statistical analysis in a few hours on this) It really reinforces the idea that having a top sprinter and sprint train is not worth it (granted we don't know what teams made leadouts, how well they worked etc which are big things to know).
For Degenkolb, Swift, Ewan and Cav to combine for 12 results out the Top15 is ridiculous (surely Ewan wasn't in Test 3?!, and something happened to Cav in Test 3 as well), whilst Ahlstrand and Grosu don't share even one.... it makes no sense! It's just flat broken.
Something is definitely up with Swift in particular although i could not say why, his stats don't have a major weakness when compared to say Cav.
Again, more specific number analysis coming later today. I think the success-failure-outside cutoffs should be different to SotD but the basic outcome remains the same. I think all should be looking to finish in the Top5 (and yes i know this is 7 guys) Grosu is the most consistent then Ahlstrand, Cav, Coquard-Swift-Degenkolb and Ewan bringing up the rear (genuinely, Ewan's + Swifts results in these tests are completely insane and i can't figure out why (apart from Ewan missing all of Test3)).
I agree with TMM that Grosu seems to be the most consistent. This is reflected in the individual rankings for this season. He has been consistent throughout the year (Barbados was actually his worst race finishing outside the top 10).
Could stamina have become really important in sprints? Grosu and Alstrand have 74 while the others have 73. One stat shouldn't really make that much of a difference though, but it's the only thing where they seem better than the rest in terms of stats.
SotD: Startlist same in all races, so Caleb Ewan was just unbelievable bad in test 3
Croatia14: I understand that it would be great with Screen Shots of trains from 5-3-1km but just 3D racing this at 8X takes a long time, so even though it was my original intention had to drop it.
If I find the time, (won't garanti at this point as things are changing in my live) I will try to create a test on only one stage with a completely straight 10 km finish X 4.0, on a very flat route to take out any MO related issues.
Taking screenshots at 10 km / 5 km / 3 km / 1 km / finish. To make a more perfect test. (guess I willl stil use this startlist as i is representing the top sprinter fairly well) Guess 3 races would be enough to get a fine picture.
TheManxMissile wrote:
For Degenkolb, Swift, Ewan and Cav to combine for 12 results out the Top15 is ridiculous (surely Ewan wasn't in Test 3?!, and something happened to Cav in Test 3 as well), whilst Ahlstrand and Grosu don't share even one.... it makes no sense! It's just flat broken.
Again, stop looking only at the riders itself but start looking at the finale helpers. It's no surprise that Viviani took so many stage wins irl but none at Cofidis. Looking at all the good sprinters they almost unanimously have trains that make no sense.
Festina f.e. has a useless set-up if they have Stauff and Petit together cause they wear each other out way too early. Then it's obviously no surprise if Coquard is in the wind too often too early. Other combinations are also tough cause Petit doesn't work with any of the leadouts (besides Daniel) and I can only see the Weber/Stauff combination as a working leadout, but even that one looks less then ideal with both being low on Res and Flat. Spotify in comparison has a well working system if they use it in a manner like Ariesen - Roelandts/Stallaert - Ahlstrand f.e., as those riders don't contradict each other. Aker should have a really good leadout if they use them in an Oss - Dakteris - Swift way or similar. P-A could be great if they use it in a manner like Matthews - Keukeleire - Ewan, but f.e. Nelson would break every train they have. But that is very logical, looking at the Acceleration stats, as Ewan uses up the same energy for the acceleration as Nelson does, hence loosing energy while his leadout increases his speed and then runs out of speed too early. That is logical, but something that sadly PCM15 didn't understand. A leadout is not just the best sprinters put together as it was in PCM15, but it's actually something that has to be very carefully composed in PCM18. I don't mind that, but if most of the people do that's okay.
Still, you can have some tough luck with the AI picking other riders than you intend for your leadout. I experienced that with Per always being used to chase the break down instead of being the third man in the train. But at least that meant that Kump didn't start from too far at the front in that race, but from too far at the back cause my then 3rd rider just wasn't strong enough. Again, tough luck for me, but composing the train right can ensure that you are less likely to have your main sprinter run out of energy too early.
At least that's my take from observing the PCM18 engine since MG started, so that's how I will adjust this offseason. Of course this is a subjective point of view, but I'm quite confident that these changes may benefit the consistency of sprints.
Of course, the faster your leadouts are the higher is the chance that your main rider wins from the perfect position. But at the same time the chance is higher that your main sprinters blows up big time. It's a trade-off of increased winning chances (I guess) vs. better consistency. But there is a reason that it was Mörkov or Richeze leading out Viviani and not Fabio Jakobsen.
Croatia14 wrote:
Yes I do. I don't think it's bad that Ewan, Silvestre, Kennaugh and Ahlstrand perform equal/better, in fact the high flat sprinters were just massively underrated in PCM15. Also it's down to the fact that the top sprinters just don't have good sprint trains right now, because they didn't need them. They don't perform as dominant as in PCM15, but I think that's very good.
I think the composition of a good sprint train is the key to success for those teams, especially one that doesn't wear their main sprinters out. Teams with Top sprinters now need to emphasis their team more on those, which is a good thing imo.
If I don't have a good sprint train, then how do you make those? I just don't understand it...
For Bryan Coquard - who has an immense sprint/ACC combo of 83,5 AVR - the best in the game this is his leadout setup:
Usually he also have a rider like Kortsidakis (74FL) and Karatzios (76FL) with him.
Riders like Degenkolb have 76FL and perform equally poor as Coquard, so I can't quite grasp why Ewan (75FL), Grosu (73FL) and Ahlstrand (74FL) should perform better than him, Swift (74 FL) and Coquard (73FL).
When a rider like van Stayen is among the favorites, he doesn't perform better either with his 76FL and help from riders such as Eislers and Dzamastagic.
So just how do you explain what is happening? Why is Grosu the most consistent flat sprinter? Why is Groenewegen perhaps the strongest this season?
Honestly I don't see a pattern that I can follow, other than I should pursue a bunch of worse sprinters next season than I have now. If it was a matter of flat stat then fine. I could just train Coquard to 75FL and then things would be solved.
Caleb Ewan, Ahlstrand, Groenewegen and Degenkolb/Swift present similar MO stats, here Coquard is slightly worse, but if I could see that Bewley and Kump was much better than usually I could accept that I probably needed to train Coquard in MO - but for now that seems to be just relevant in terms of getting into the line with the bunch and not so much as an effect on the sprint itself.
Maybe we're also expecting too much from the game with this DB? I mean we can tweak corners and roads a bit, but at a certain point we'll just have to accept there are limits and it won't get any better.
Croatia14 wrote:
Yes I do. I don't think it's bad that Ewan, Silvestre, Kennaugh and Ahlstrand perform equal/better, in fact the high flat sprinters were just massively underrated in PCM15. Also it's down to the fact that the top sprinters just don't have good sprint trains right now, because they didn't need them. They don't perform as dominant as in PCM15, but I think that's very good.
I think the composition of a good sprint train is the key to success for those teams, especially one that doesn't wear their main sprinters out. Teams with Top sprinters now need to emphasis their team more on those, which is a good thing imo.
If I don't have a good sprint train, then how do you make those? I just don't understand it...
For Bryan Coquard - who has an immense sprint/ACC combo of 83,5 AVR - the best in the game this is his leadout setup:
Usually he also have a rider like Kortsidakis (74FL) and Karatzios (76FL) with him.
Riders like Degenkolb have 76FL and perform equally poor as Coquard, so I can't quite grasp why Ewan (75FL), Grosu (73FL) and Ahlstrand (74FL) should perform better than him, Swift (74 FL) and Coquard (73FL).
When a rider like van Stayen is among the favorites, he doesn't perform better either with his 76FL and help from riders such as Eislers and Dzamastagic.
So just how do you explain what is happening? Why is Grosu the most consistent flat sprinter? Why is Groenewegen perhaps the strongest this season?
Honestly I don't see a pattern that I can follow, other than I should pursue a bunch of worse sprinters next season than I have now. If it was a matter of flat stat then fine. I could just train Coquard to 75FL and then things would be solved.
Caleb Ewan, Ahlstrand, Groenewegen and Degenkolb/Swift present similar MO stats, here Coquard is slightly worse, but if I could see that Bewley and Kump was much better than usually I could accept that I probably needed to train Coquard in MO - but for now that seems to be just relevant in terms of getting into the line with the bunch and not so much as an effect on the sprint itself.
You're only looking at FL and not RES/STA which are also very important stats when working high endurance efforts, which leadouts are. I don't have time to look at the db right now to look at those matches, but support stats do play a role.
I havent looked in much detail at the tests myself but based on SotDs summary of the results for individual riders, i have to agree with him. That looks awful.
whitejersey wrote:
You're only looking at FL and not RES/STA which are also very important stats when working high endurance efforts, which leadouts are. I don't have time to look at the db right now to look at those matches, but support stats do play a role.
Coquard have 73RES which is the highest of all top sprinter except for Bewley and Guarnieri. Stamina in the past only mattered when moving outside the 200km zone. Is this different to previous versions? Anyway, Degenkolb and Swift have 73END whichi is similar to Ewan, Ahlstrand and Grosu (73/74).
My leadout sprinters have 74END, 69RES (Stauff) and 76END, 72RES (Petit). Weber also have 74END, 69RES while Barbier is 71/71.
In comparison a rider like Roelandts (who is supposedly a good leadout rider) is 73FL, 72END, 71RES.
I don't buy that leadouts have to be pretty crap at sprinting as it really should be meassured up against the lead sprinter. Obviously you don't want too similar riders, but if 79SPR 77ACC is good for Ahlstrand who is 83/82, then surely 80/78 ACC is similar to Coquard who is 84/83. The gap is exactly the same.
In PCM 15 the highest sprint/ACC wasn't that important. It was however important to have leadouts with high FL, which is why I had riders such as Emerson Santos working as primary leadout for Coquard. 78/78S PR/ACC. I see absolutely no indications that it works that way in this season. Coquard performs just as random when he have Stauff as leadout as he does when he isn't there.
In Tirreno-Adriatico Coquard won a sprint and ended 4th in the other. But I don't think that is representative, as his strongest opponent was Groenewegen and Nizzolo so him performing "decently" was pretty much a matter of him being superman compared to the others. Yet he performed only a 4th in the final stage despite of his 84 recovery stat.
In Ronde van Nederland he has the same setup around him, which on paper should be the strongest if I conclude from croatias posts. There he was 16th, 6th and 9th. So not particularly impressive to say the least!
So while I get the point that it is about finding the proper setup to lead out the sprinters I still struggle to see any team getting that one right tbf. I have spent quite some time sampling and is using this season as some sort of experiment to see what works best, but in honesty nothing works - for anyone. Except those that doesn't have a train.
The late hill at stage 4, world often allow a minor split so a true sprinter stage best without late climbs (even small), if we do not lift low MO
Seemingly good fair sprints with a lot of room for moving and changing position, often very close photo finish.
If that small bump towards the end of the stage is ridden in a way that MO instead of hill then the game design is imo completely ridiculous and shouldnt have ever be used for the mg
Tamijo wrote:
SotD: Startlist same in all races, so Caleb Ewan was just unbelievable bad in test 3
Holy fuck! (yes it's uncensored swearing time lads)
I....
what?!!!!
Just so i can get some context in the wider game: is there a Puncher, Stage Racer, Climber, Classics rider that just completely vanished from the Top15 for an entire Stage Race that was perfect for them? Who is also in the Top5 Individual ranking for that specialization?
I genuinely don't know the answer. And ok this is 1-in-5 times, but that's an insane percentage to just vanish! It is completely game breaking!
(this also means Cav utterly vanished for half that test 3, i.... just... what!)
Croatia14 wrote:
Again, stop looking only at the riders itself but start looking at the finale helpers.
I can't look at this, as we don't get that info for the tests Tamijo was kind enough to run.
P-A could be great if they use it in a manner like Matthews - Keukeleire - Ewan, but f.e. Nelson would break every train they have. But that is very logical, looking at the Acceleration stats, as Ewan uses up the same energy for the acceleration as Nelson does, hence loosing energy while his leadout increases his speed and then runs out of speed too early. That is logical, but something that sadly PCM15 didn't understand. A leadout is not just the best sprinters put together as it was in PCM15, but it's actually something that has to be very carefully composed in PCM18. I don't mind that, but if most of the people do that's okay.
Just picking on my team bits here of course.
Can you explain to me how ACC works in PCM18? I think i must be working on the wrong idea, because in PCM15 i could (and did) use Ricki as a leadout to success on multiple occasions. Although i normally won't use him as a leadout but opt for Kek for basically what you state in that Ricki is quite strong for a last-man.
The problem with PCM, not just 18 but all PCM's going way back to 10, is that the AI does not form good trains. Too often i will send Matthews, or Edmo, or Alaphilippe, as that 3rd-man only to see the AI using them to relay and then try and use them as a lead out as well. If at all.
Granted i had a lot more experience with the older PCM engine from around 12-15 and knew how to build that team up. But when, on numerous occasions, i asked about changes for 18 the response was always the same, sprinting has not changed apart from MO being more valuable.
Now having played some more PCM19 lately, i still form my trains manually in the same way i did on 12-15, and get the same result. I'm just no seeing the differences that aparently are there? It's frustrating if i'm doing less well from something i just do not know about.
(yes i know Ewan is high in the rankings, mostly due to sweepeing a few U25 jerseys along the way)
Of course, the faster your leadouts are the higher is the chance that your main rider wins from the perfect position. But at the same time the chance is higher that your main sprinters blows up big time. It's a trade-off of increased winning chances (I guess) vs. better consistency. But there is a reason that it was Mörkov or Richeze leading out Viviani and not Fabio Jakobsen.
So my plan to stack more tier-2 sprinters at the expense of a lead-out system could actually be a good idea. I am happy to stick with this plan for a season and see how that works in the PCT with more flexibility in race selection.
Tamijo wrote:
SotD: Startlist same in all races, so Caleb Ewan was just unbelievable bad in test 3
Holy fuck! (yes it's uncensored swearing time lads)
I....
what?!!!!
Just so i can get some context in the wider game: is there a Puncher, Stage Racer, Climber, Classics rider that just completely vanished from the Top15 for an entire Stage Race that was perfect for them? Who is also in the Top5 Individual ranking for that specialization?
I genuinely don't know the answer. And ok this is 1-in-5 times, but that's an insane percentage to just vanish! It is completely game breaking!
(this also means Cav utterly vanished for half that test 3, i.... just... what!)
George Bennett 37th in Catalunya in a field which he entirely beat later in Japan (barring Uran and Tenorio). Both 6 day MO/HI races.
24/02/21 - kandesbunzler said “I don't drink famous people."
15/08/22 - SotD said "Your [jandal's] humour is overrated"
11/06/24 - knockout said "Winning is fine I guess. Truth be told this felt completely unimportant." [ICL] Santos-Euskadi | [PT] Xero Racing
Testfile
4.0 pancake flat 172 km. No turns in last 10 km.
No garanti it will be different, but from a stage profile point of view we need to know not to use many hours if it is to no use, can't eliminate daily form but otherwise this should be a perfect flat route.
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
