It's clear to me that the game is shrinking. It's terrible, but it's true. It's at its best when we have 3 full divisions flowing well between, but those days are over. To me, we have two options. We can do some intense advertising in other places to try to attract applications to the MG, which is a first step I'd probably prefer, although it may not lead to members that are especially committed to the forum. Or, we can accept the reality that the game needs to be condensed in some way.
The easiest solution is simply to condense PCT. It's the largest division and while it might be the most fun as well, keeping it that large is unnecessary imo. Putting 4-5 PCT teams in CT goes a long way towards increasing the solvency of three divisions, at least in the short term. It doesn't change the fact that the manager count decreases yearly, but it gives us some room to breathe on that front.
Of course, the problem is that it's impossible to do this fairly. We can't tell promoting CT managers that, "oops, no promoters this year." That's completely undoable. The only fair way to do it is simply not promote anyone under 5th in CT due to disbandments for a couple of years. We never hope for disbandments, which is why ideally this should have been done last year already. But even if it takes 2-3 years to shrink PCT to the proper level, I think it's worth it.
Another option includes moving to two divisions. This is a last resort to me. I love the game with 3 divisions, and it would require massive restructuring to reduce it back down to 2 a full decade after expansion. I don't think the game could survive that kind of change at this point. Not to mention, we still have too many CT managers at the moment for that to really work without a complete calendar and race day allocation overhaul. Which is a good problem to have!
Finally, we could go back to amateur teams. I enjoyed participating in amateur teams, but honestly I don't think they're great for the game. I'm sure we could fill CT up with plenty of managers willing to take on a second team, but is that really what we want? We'll end up with a 1/4 of managers having 1/2 of the teams in the game. Doesn't seem right for RP or the market.
On Game Choice:
This kind of a damned if we do, damned if we don't thing. PCM is a poorly made game. We've already made so many adjustments to the DB, stages, etc. already that a return to PCM 15 doesn't seem viable to me anymore. That's a shame, cause I still do prefer it for our type of DB and game, but I think that's the reality. Plus, don't forget that we had a good number of people who were ready to quit if we went back to 15 last year.
At the same time, Cyanide continue to put out games that are worse and worse AI, at least for our purposes of needing races no user interraction and an inflated DB. I'm not sure there's a noticeable difference between PCM 18 and 20 for me at this point. I haven't played 20 though, so I would need to see testing. But they seem pretty similar with a few tweaks: fewer beakaway wins in 20, a bit better mtn AI, but even more potential for weirdness. Sprints are still broken and TT's are not good.
I think once we moved on from 15, we basically made the decision that we're just gonna go with what PCM gives us. So honestly, I'd be looking towards using 21, though I'm definitely not optimistic there'll be improvements. That being said, if we're going to lose long time managers because of a specific game, it's never worth it. Plus, there's the reporter issue. The newer game we play, I think it limits the potential reporters. I know for one, I could run PCM 15, I could barely get through a stage in PCM 18, but PCM 20 is outside of my computer's range. So I can only offer previews now. (Not that I was ever the most prolific reporter tbf).
Personally, I'll continue no matter the game. I'm in too deep to ever leave by my own choice.
On Wildcards:
I'm definitely sympathetic to PT managers who see riders without any stakes taking major points from them in PT races. Obviously Guldhammer was a rider like that recently in the Giro. Obviously I like it the way it is, but I wouldn't be opposed to it working more like CT in HC, where it does use race days and you can score. I mean, if I had put Guldhammer in the Giro and he COULD score, it would still be more than worth it. He probably would have scored more than the rest of his point total for the whole year there. I think it would be worth the risk for a lot of PCT managers for that reason - it's usually the very best on a terrain from PCT that seek wildcards in PCT anyway.
There would need to be a totally new system though because obviously as is, only 6 PCT teams can ride GT's, and they'd get a huge benefit over those getting lesser PT races points wise, more than the current money difference. Not enough PT race spots to go around either, in any case. Maybe GT's would cost more racedays or something, but it would be a whole thing.
On sprints:
I don't think much has changed here. We've put up with broken sprints for a while now. I disagree with quadsas's assessment here. A lot of it is random, race to race, year to year. AKA doing better in PCT and PT than he ever did in CT with the exact same stats doesn't make any sense whatsoever. Reinhardt being a beast last year na d a nobody this year is the same. Ewan shouldn't be behind half a dozen other sprinters in PCT. I think that's obvious, and it's not "TMM's fault" that the best rider is far from the best despite ample support. It's not about every sprint having an exact perfect ending with regard to the relevant stats. It's just that a rider shouldn't win one day and then literally not even join the sprint the next day, or vice versa.
The proof is in PCM 15 races, cause that's where sprints were perfectly fine. There would be oddities of course, mostly due to inflation, and there would be variation stage to stage. But the best sprinters always sprinted, they did better with good leadout trains, and daily form or particular race dynamics had to be the difference makers, not a bizarre energy system or weird AI decisions. The results AND actual action made sense. And the roleplay aspect is a huge part of this, so even if the end of the year rankings do even out, the races matter as much.
At this point though, I'm not sure what works to fix it. I think TMM's fears about adding mountain stages to the end are accurate. It would fix a big problem but could start a bunch more. We'd have to do testing for that.
I really like cunego's idea of reducing the gradient of finishing sprint to -1-2%. We see on the few downhill finishes that the energy bonking problems don't happen as much. However, not sure it's feasible. As redordead noted, it would take an extraordinary amount of work to perfect each stage. And it would HAVE to be perfect. If the slopes go too low, you create the opposite problem where nobody can pass, and you introduce other stats like downhill when they shouldn't apply. It's a very precarious solution, but I do think it has merit.
Oh, and let's not forgot that the biggest problem with sprinters, for results oriented folks, is that they're impossible to make sense of in the market now. Guys demand 150k wages when 50 similar riders can be picked up for 50k from FA. Meanwhile, we adjust OVL's for what we theorize could be the differences, and the next year the results are the opposite. Completely untenable situation.
On re-runs:
I'm fine with reporters rerunning the race if AI is really out of whack. We have a very good crop of reporters who I trust entirely to do the right thing. And practically, they already have this power and could be doing it the whole time, and we'd never know. So I don't think a tempered endorsement is going to change anything except remedy some insane stages.
Not to mention, we accept similar reruns already. For instance, I reported Roubaix once, and I had to replay it like 3 times because it kept crashing in the final 10-20 km and I couldn't figure out why. Some people probably lost out from those reruns, but those kinds of things we've always accepted. This should be the same.
On puncheurs:
I think puncheurs races have been remedied for the most part by the stage adjustments after 2019. A lot of good work was done, and puncheurs can win hilly stages again. The biggest problem are hybrid races where puncheurs are supposed to have a chance. They simply can't compete with the time gaps climbers can create on even mild mountains. That might be fine irl, but it's a big issue with our DB and goals.
I really like the idea of giving hilly days bigger/more time bonuses in stage races that should favor them, just like sprinters. I think that's a great idea that serves the same purpose for a similar problem. Are there any downsides? I can't think of any.
Other stuff:
Maybe not the thread or place, but let's not forget about other issues: particularly the DB inflation, method/number of talent additions yearly, and progression path adjustments. These also need to be addressed for the game to thrive, and every year we kick the can down the road, making the future less and less viable. And there are plenty more I'm missing. Those need a spot in this discussion too, while changes can still be made.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
I don't think much has changed here. We've put up with broken sprints for a while now. I disagree with quadsas's assessment here. A lot of it is random, race to race, year to year. AKA doing better in PCT and PT than he ever did in CT with the exact same stats doesn't make any sense whatsoever. Reinhardt being a beast last year na d a nobody this year is the same. Ewan shouldn't be behind half a dozen other sprinters in PCT.
He isn't, though, at least not from a ranking perspective. He's behind Silvestre, Manninen and Halvorsen, yes, but those are actually exactly the type of sprinters who have been successful over the past two years: high flat/res combo with at least 80 Sp/Ac and no leadouts. Top sprinters in PT? Gaviria and Kemboi, highest Fl/Rs and, if I'm not mistaken, rarely relying on a train. Jakobsen very neatly fits that definition, too. This doesn't come out of nowhere.
In that way, I actually agree that sprint success is not entirely random. Ewan's currently 19th as the 4th best sprinter in PCT, he was 24th as the 4th best sprinter in PCT last year and 21st as the 3rd best pure sprinter in PT the year before. Top sprinters like Grosu, Kennaugh, Degenkolb and Coquard have consistently placed in the Top 30 in PT. We have outliers both ways like Groenewegen overperforming two years ago or someone like Howard maybe underperforming a bit last year, but that's normal. There's always going to be variation, and it would be boring otherwise.
Reinhard's case is on paper extreme, but I think that's one part simply regression to the mean as far as racing against the same competition goes - he was just bonkers last year, and if he hadn't performed to his absolute limits then, we likely wouldn't talk about it much now -, and not being considered anywhere near a favorite for PCT (71 fl, 79 sp will not do, even with that great acceleration).
His 260k salary is a good case though that sprinters' expectations in renewals need to be reworked, even when taking results into account. And obviously it's a shame that his underperformance coincides with Quintana's on the same team. I completely understand the frustration, and I really hope that some good results come around that might sway you to stay in the game, Fabianski!
So while I do think it's worth trying to optimize the sprinting experience, I don't think it's fair to say that there is complete randomness, neither from race to race nor from game to game.
I don't think much has changed here. We've put up with broken sprints for a while now. I disagree with quadsas's assessment here. A lot of it is random, race to race, year to year. AKA doing better in PCT and PT than he ever did in CT with the exact same stats doesn't make any sense whatsoever. Reinhardt being a beast last year na d a nobody this year is the same. Ewan shouldn't be behind half a dozen other sprinters in PCT.
He isn't, though, at least not from a ranking perspective. He's behind Silvestre, Manninen and Halvorsen, yes, but those are actually exactly the type of sprinters who have been successful over the past two years: high flat/res combo with at least 80 Sp/Ac and no leadouts. Top sprinters in PT? Gaviria and Kemboi, highest Fl/Rs and, if I'm not mistaken, rarely relying on a train. Jakobsen very neatly fits that definition, too. This doesn't come out of nowhere.
Small point sure, but as of the last rankings update, you forgot AKA is ahead of him too, so Ewan is 5th of the pure sprinters. That actually supports the high flat/res theory though, for sure. Though I hold that Reinhardt outscoring AKA by over 200 points in CT last year, despite being the exact opposite archetype, shows that there is a significant degree of randomness and unpredictability still present.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
cunego59 wrote:
the type of sprinters who have been successful over the past two years: high flat/res combo with at least 80 Sp/Ac and no leadouts.
Real life: Let's build a sprint train, then we'll have high chances of succeeding.
PCM20: Managers praying that the AI won't build a sprint train with their team...
Btw, Reinhardt worked pretty much the same last year with and without train, i.e. sometimes pretty bad (or even not participating), but mostly he did well. And with the train often tearing things apart, he rarely got a really bad result when we did.
Obviously, he overperformed last year. And probably Acc was more important back then than it is with PCM20. But the comparison with AKA is a pretty good one - why would two sprinters with exactly the same stats and pretty much the same setup (leadout for Reinhardt, none for AKA) have pretty much inversed results in back-to-back seasons?
And of course, the wage issue needs to be looked into. Reinhardt's 260k are ridiculous given his performance, but part of it definitely was bad negotiation strategy. I don't know how many 50-100k sprinters have scored more points tbh, but I guess there are quite some of them.
Can I just point out that Bonifazio would also fit the "high flat/res combo with at least 80 Sp/Ac" category, perhaps more than anyone? And while he has improved this year, he's still not one of the best in the division. And don't get me started on last year (where he had the exact same stats by the way).
EDIT: I got triggered. He scored 76 points the whole season! 76!!!!!!!!!
cunego59 wrote:
the type of sprinters who have been successful over the past two years: high flat/res combo with at least 80 Sp/Ac and no leadouts.
Real life: Let's build a sprint train, then we'll have high chances of succeeding.
PCM20: Managers praying that the AI won't build a sprint train with their team...
a) Yes, that's kinda stupid, but it's what we're working with
b) I never had to worry that my team would build a sprint train. That's not entirely out of our control.
Fabianski wrote:
But the comparison with AKA is a pretty good one - why would two sprinters with exactly the same stats and pretty much the same setup (leadout for Reinhardt, none for AKA) have pretty much inversed results in back-to-back seasons?
This can be seen in different ways, I suppose, but I don't think it's necessarily contradictory. You have a bunch of very strong riders with similar characteristics who perform relatively consistently well. And you have a ton of other riders who are capable of good results but can also disappear for lots of stages. Guys like perhaps AKA (because of poor acc) and Reinhard. (And true, I did overlook AKA. My point still stands though, I think.)
If you're in the second category, then yes, there's a good bit of randomness. But also, while for most riders in that category, good and poor results maybe roughly even out, there will be over the course of the season necessarily some riders who accumulate disproportionately many good or poor results, respectively. Our "job" is to try and parse which good seasons are stable (as I've tried to point out, there are sprinters who are stable) and which are more likely the result of random accumulation of good results (which can then, in extreme cases, completely flip either way). That's tricky, but I would argue that at least to some extent, it can be done.
@Ollfardh: Bonifazio does perform pretty well this year indeed, as he should. I honestly don't know what happened last season, but iirc, you did bring him to a lot of races together with either Gaviria or Mareczko, no? And not so much anymore this season? Correct me if I'm wrong. That might be one part of the explanation.
€dit: Just to make it clear, I don't want to argue that sprints are *fine, actually*. And I know that maybe I can talk because I have Silvestre and he's doing what he's doing. Again, I think we should try to make adjustments. But I also think there's an important distinction between specific issues that we have (trains, GC sprinters ...) and the general assumption that everything's just completely random anyway. I'm just laying out my thoughts on why I think the latter is overblown, or at least contained to less-than-elite riders, which I think is acceptable and likely immutable anyways considering our database.
Edited by cunego59 on 28-01-2022 19:23
Fabianski wrote:
Real life: Let's build a sprint train, then we'll have high chances of succeeding.
PCM20: Managers praying that the AI won't build a sprint train with their team...
"Real MG life": I always hope for a lead out because Saber does better then. Unfortunately in a stacked PCT field he doesn't get it often.
Fabianski wrote:
Btw, Reinhardt worked pretty much the same last year with and without train, i.e. sometimes pretty bad (or even not participating), but mostly he did well. And with the train often tearing things apart, he rarely got a really bad result when we did.
Did he really? From my memory he worked great as an underdog in HC/C1, but often disappointed you in C2HC/C2. Maybe I'm wrong.
Fabianski wrote:
And of course, the wage issue needs to be looked into. Reinhardt's 260k are ridiculous given his performance, but part of it definitely was bad negotiation strategy. I don't know how many 50-100k sprinters have scored more points tbh, but I guess there are quite some of them.
I also think some issues regarding renewals should be fixed, although some improvement were made. The fact remains that the demand will always come from previous season's results.
alexkr00 wrote:
About the GC leader bug. I agree it looks fishy. I took a quick look at some flat races this year and the results definitely seem to suggest it. But, from what I can see there were other sprinters also not sprinting in all of these scenarios. And because of this while the evidence looks overwhelming towards the bug, I'm not sure if there's really a bug or a whole awful lot of coincidences.
Spoiler
Race
Stage
Leader
Sprinted
Was the only big sprinter missing?
Tour of Qatar
3
Gaviria
Yes
Tour of Qatar
4
Gaviria
No
No
Tour of Qatar
5
Gaviria
No
No
Tour of Qatar
6
Gaviria
Kind of (top 10 finish)
Baltic Chain Tour
2
Saber
No
No
Baltic Chain Tour
3
Silvestre
No
No
Baltic Chain Tour
4
Van Aesbroeck
No
No
Baltic Chain Tour
5
Jakobsen
No
No
Circulo de Juarez
3
Maninen
No
No
Circulo de Juarez
4
Bouhanni
No
No
Circulo de Juarez
5
Howard
No
No
Barbados
2
Silvestre
No*
No
Barbados
3
Ewan
No*
No
Barbados
4
Bonifazio
No*
No
* seemed to be there when the trains were made but did not seem to get involved
I should find my post, i think it was in Juarez discussion, where i went thrugh these races and showed the difference between riders in a "normal" stage, then when they were Top3 GC. Every single example, bar one (and i explain that case as well, the Top3 GC result was significantly worse.
This is a repeatable pattern in multiples divisions, and was also a problem going back to PCM18 (i've checked).
Spoiler
In fact i've gone further with this and looked back over all those reports to use the screenshots and commentary to try and figure out more about this.
I believe, very strongly, that i am correct to say these riders are given a "GC Leader" tag by the game AI. Because they follow the same behavior patterns as a normal GC leader would in a normal GC race.
I.e. They stay fairly forwards in the pack for the day, avoiding crashes (AI defeinitely doesn't know they are turned off), getting shelter, right side of gaps etc. This is until the finish. As a GC Leader they don't get involved fully in the sprint to reduce the crash chance (again, PCM AI isn't that smart guys). However the game can recognise these are strong flat riders and can get solid result. So they then start to participate, and because they are still good sprinters, gain places reasonably fast at the end.
This bumps their finish position slightly from a normal GC leader, but still way off the stage win. In fact this also explains some of these riders best results coming on stages where the sprint breaks up, and gaps appear between groups, as the GC Leader AI panics to close down and defend the GC position.
Granted this is second hand data as i didn't play the stages. And as far as i know this isn't someting visible in a DB file we can use as proof. But to my eyes it's such a clear explanation of the "Bug".
Some sprinters will always miss out, because our MG DB is inflated beyond broken. We can't fix that without seriously radical solutions. And that's fine, as long as it's a dice roll and not because you're Top3 in the GC.
One last thought on the GC bug, also something that has to be tested, but it might be an option to edit the standings after the first stage and give the 3 control team riders like 20 minutes of an advantage on everyone else. Then, no sprinter could be held back by a Top 3 GC rank. Now, obviously that might bring a whole host of AI issues (-> testing), but maybe not as much. I can recall plenty races where breakaways have been caught easily even with the yellow jersey team not working. This would require some effort from reporters, both for editing and controlling the control team to make sure they don't behave like the yellow jersey team, but if it's contained to sprinter stage races, it may not be too bad.
cunego59 wrote:
b) I never had to worry that my team would build a sprint train. That's not entirely out of our control.
I guess once everyone stops trying to do trains (i.e. sending a decent leadout), it will be pretty random. That's clearly not the goal, but I guess it's inevitable. Maybe it would be good for breakaways, though ^^
cunego59 wrote:
You have a bunch of very strong riders with similar characteristics who perform relatively consistently well.
I'm just not quite able to figure out what it takes. This year it looks like high flat is an advantage, true. But what do I do with Howard or Ewan, who have higher flat than e.g. Manninen or Jakobsen?
Is it high Res? Then Silvestre shouldn't do quite that well, as 71 isn't *that* great.
I'd expect Spr to be the most important value, but it is clearly not. But even if it isn't, Ewan should perform way better than he does (same goes for Howard). So I guess even if you get one of those "very strong riders with similar characteristics", it's still quite a bit of luck that you need. But that's my opinion.
And I fully agree that Reinhardt shouldn't be among the top contenders in such a field - but he was last year. So why not this time around? As I said, I guess Acc has lost quite some importance (which might also be a reason for AKA to do much better now) - which unfortunately wasn't clear pre-season. Else I'd never have renewed him...
Fabianski wrote:
Btw, Reinhardt worked pretty much the same last year with and without train, i.e. sometimes pretty bad (or even not participating), but mostly he did well. And with the train often tearing things apart, he rarely got a really bad result when we did.
Did he really? From my memory he worked great as an underdog in HC/C1, but often disappointed you in C2HC/C2. Maybe I'm wrong.
Well, he had a stage win in Vancouver (C2HC) and won two stages and GC in Amissa Bongo (C2), complemented e.g. by two 2nd places and 4th overall in Solidarnosc (C2HC). It's true that there were some disappointments (mainly Ras Tailteann, where for whatever reason Cissé sprinted on stage 1 and Reinhardt just didn't care all race long), but he was mostly way inside the Top 10 when not suffering from "GC curse".
C1/HC results were in general slightly lower, but still far better than they are now. Funnily, I guess he did his most consistent race so far this year in Qatar
1. I am pretty optimistic about the MT strategy. I don't think being having the GC favorite in a MT race changes behavior much in sprints. You may be less likely to go into breaks but breaks in sprint races are pretty useless. If you only send 6 riders that might be an issue as I have seen sprinters not sprint when they have a GC leader and only 6 riders on some stages. So I would include that in the testing. Making the control rider high MT might help with these issues.
2. I am afraid creating down hill finishes is not realistic - that is a lot of work. Also, in my experiences up hill finishes actually have the desired effect so not sure it is the right solution. Also, making all the finishes wide and straight a few years back probably made it worse, so this would need testing.
Unlike the GC issue I am not optimistic about fixing the lead out problem (the sprint AI is different in classics, so it doesn't surprise me that they work better). The frustrating thing is if the AI started the sprint just 300 meters later it would be fine. I do think that it is worth testing 21 to see if Cyanide did anything.
To repeat my favorite point, culling the database of excess sprinters would help and is much more important than in other disciplines. Trains aren't as bad when you don't have 20 equally skilled sprinters behind you (See Keough's stage win in ToC). In Baltic Chain the top 20 sprinters in SP+AC were all no more than 2.2% away from the average of the 20. This should generate apparently random results over any small sample and likely exaggerates any AI issues.
Game Version
While I agree 18 sprints were a little better they still had the the GC problem. I think TT's were also pretty good in 18 but are weird in 20, particularly the mid-distance ones. That said I would not go back to 18. I think 18 had way too many breakaway wins and I would much rather see guys like Areruya and Dombrowski getting stage wins than many random breakaway riders. And I don't think the trade off is worth the issues caused by going back. I also think 20 does have more stage to stage variation which can keep it interesting but may explain some really bad performances in races with only one or two mountain stages.
If we go back I would go back to 15 as my impression is it worked well for the mangame. I am not that worried about stages, there are a lot of stages available for 15 and the mangame doesn't need endless new stages.
As SotD noted in another thread - there is merit in a predictable AI even if it is flawed. One way to help with predictability is to stick to a version for a few years (contrary to my point about 21 above). While I think some of the Rheinhardt thing is luck, I think 18 did value ACC in sprinters a little more so unbalanced riders like him and AKA see an impact that might otherwise not be discernible. So Fabianski got the double whammy of an OVL increase that valued ACC and a game AI that doesn't. Sticking to one version helps this kind of issue.
Divisions
Finally, the game would have to get pretty small for me to vote for 2 divisions. With 2 divisions 1/2 the players are struggling at any one time, with 3 it is 1/3. The latter is much more fun.
I think its pretty apparent that sprinters are by far the most inflated class of riders in the game and without culling them we will probably never find a setup that works correctly, when close every other aspect of the game is relatively ok. I understand the frustrations of some people, I must admit that I don't remember how 15 sprints worked but I do know that it was absolute ass at handling hills depth.
When you have 22 sprinters in the CT alone that have more than 79 SP as a stat it is bound to cause issues if you expect them all to perform, that's more than 1 per team. You're bound to have spikey performances from them. This type of congestion will never allow the AI to fully work with what its got because the MG db is a broken mess, that we all love, but I think that more than anything else the sprinter area is probably the one that needs the most reworking.
To put it into perspective, there are just 3 climbers in the division with more than 79 MO.
I cannot attest to anything with TTs since I avoid them like the plague.
Fabianski wrote:
It's true that there were some disappointments (mainly Ras Tailteann, where for whatever reason Cissé sprinted on stage 1 and Reinhardt just didn't care all race long), but he was mostly way inside the Top 10 when not suffering from "GC curse".
While I do agree that sprinter inflation is a thing and a problem, I do want to ask, are we sure the alternatives are better? Specifically, if we want to get the number of top sprinters down to maybe 12-14 per division, and then through race days and clashes try to get the number of competitive sprinters per race to 8-10 (or even fewer), then around half of the teams in a race like Juarez or Barbados have ... nothing at all to play for?
I don't have either a top puncheur or a top climber, but in those stage races, I can at least hope for KoM points, breakaway wins are much more likely, too, and second tier riders have a clear pathway to at least backend Top 20-25 points. In sprint races, maybe strong rouleurs have a slightly better chance with fewer sprinters there, but I wouldn't be sure of that.
I guess there are always going to be racedays that are going to be wasted, and maybe I'm overthinking this - from what I can tell, it didn't always used to be like this with so many sprinters and it probably worked back then? But I wanted to bring this up still, if only out of curiosity what others think about it.
Also forgot to mention, we need to keep 3 divisions.
But the training cost should be scaled for the 3 divisions somehow, being stuck in CT without the advantage of training will continue to shrink the game imo.
cunego59 wrote:
While I do agree that sprinter inflation is a thing and a problem, I do want to ask, are we sure the alternatives are better? Specifically, if we want to get the number of top sprinters down to maybe 12-14 per division, and then through race days and clashes try to get the number of competitive sprinters per race to 8-10 (or even fewer), then around half of the teams in a race like Juarez or Barbados have ... nothing at all to play for?
I don't have either a top puncheur or a top climber, but in those stage races, I can at least hope for KoM points, breakaway wins are much more likely, too, and second tier riders have a clear pathway to at least backend Top 20-25 points. In sprint races, maybe strong rouleurs have a slightly better chance with fewer sprinters there, but I wouldn't be sure of that.
I guess there are always going to be racedays that are going to be wasted, and maybe I'm overthinking this - from what I can tell, it didn't always used to be like this with so many sprinters and it probably worked back then? But I wanted to bring this up still, if only out of curiosity what others think about it.
Having a top level puncher in a divison etc.
Wasted race days, and unexpected results have to be a factor, just like in real life. If everything is calculated before the first race of the season, then we can just put the ores back in the boat and sink the entire game imo.
The key here has to be training in some sort of form. So that if you have a 79/80 sprinter you can invest money in making his stats stronger and that way, thoose who want to increase there chances can do it through training. There are too many at the 79 to 81 range at the moment. So if the cap was at 79, and then through traning and it was cheaper to train certain stats in certian divisions, it would even it self out in a couple of seasons. But what do i know, im a CT legend at this point.
We could consider a sprint only stat inflation formula to reduce the amount of similar levels there.
It would need some tests to e.g. guarantee that riders don`t stop sprinting at all, though e.g. question is if a 77 or 78 rider should sprint on a stage with 15 80+ sprinters anyway.
Old
New
Diff
85
85
0
84
84
0
83
82
-1
82
81
-1
81
80
-1
80
79
-1
79
77
-2
78
76
-2
77
75
-2
76
73
-3
75
72
-3
74
71
-3
73
70
-3
72
68
-4
71
67
-4
70
66
-4
…
…
…
Such system would extend the gap a bit between the top top class that needs to be spent money to reach, e.g. Coquard as they basically remain or a bit lower, e.g. Ewan get reduced very slightly.
The more you go down, the bigger the change to get rid of all those 75-79 riders.
Maybe it could even be more extreme to have a bigger gap downwards like this:
Old
New
Diff
85
85
0
84
83
-1
83
82
-1
82
80
-2
81
79
-2
80
78
-2
79
76
-3
78
75
-3
77
74
-3
76
72
-4
75
71
-4
74
70
-4
73
68
-5
72
67
-5
71
66
-5
70
64
-6
…
…
…
It would be a way to "maybe" fix the biggest PCM20 issue, which might be AI related of course but also is impacted by the MG DB with the hundreds of similar sprinters, which obviously leads to randomness as others have stated the difference is often below 2 %.
Testing would be needed of course to see if such changes would lead to anything good AND the question is how to handle this in the OVL/Wage system. Obviously sprinters in general would ask for less then, which is fine I guess. But it would be needed to check if the top sprinters then, who also might have a small OVL change aren`t overpowered then if competition for them is lower and points scoring potential is rising this way.
And also things like GC leader sprinter then would need to be tested a bit if this impacts it as well due to number of top level sprinters and so on.
Just lowering sprint stats won't work, it needs to be about differences. In fact, I already think we've "improved" over the last few years. There used to be too many similar 81-82 sprinters, now it's too many 80 sprinters. If we lower everyone down to 78, it will still be the same problems. I think the only way to fix this is to add more diversity in the new riders. I feel like Gaviria was the last really good sprinter added, all the others were too low to fight this inflation problem, but that won't solve it.
We need to add sprinters that will naturally grow to 82-83 SPR (same for all other rider types btw) to make sure that we don't have 20 similar sprinters in each race. Additionaly, there's still too many obscure riders given decent stats just to add good riders from exotic nations.
Finaly, leadouts not working properly is probably the biggest problem. If the Podium Ambition train would regularly drop off Ewan in perfect position with 1K to go, Ewan would be the best scoring sprinter in the division (although I want to want everyone not to look too much at the standings with very different racedays used) and we would be having a different conversation here. Can't fix Cyanide stupid though.
With the reduction I planned, the diversity should come as it`s not just -1 or -2 for all but in a stairs way.
With training then or further stat decrease age wise the changes will be bigger as well as e.g. the new system right now wouldn`t have a 81 or 77 stat due to the formulas.
That all said, a first couple tests show two things with the 2nd system.
- leadouts still are too early or even if the sprinter only leaves at 600m the next will still pass him with 100m to go...
- Top3 GC sprinters "participate". They are always there with 5-3 km to go. But they still don`t finish as high as they should. Still lack in comparison to similar sprinters that are not in top3 GC.
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
