|
Suggestions for 2024
|
| Fabianski |
Posted on 04-04-2024 22:23
|
Grand Tour Specialist
Posts: 4667
Joined: 29-09-2018
PCM$: 185.00
|
jandal7 wrote:
Fabianski wrote:
On the other hand, puncheurs with low sprint have become rather useless on such stages (whereas e.g. Moscon won several stages ending in a reduced bunch sprint for me in the past, no way he'd do that this year for Zalgiris despite the +1 Hi training).
Not wading in on everything else, but are they getting dropped or just not winning the sprint? Because they shouldn't really be winning the sprint so I would see that as a positive for the new AI 
I'm not judging whether it's positive or negative - it just feels different from the past version. It's something managers will need to adapt to - this just wasn't really possible without knowing it beforehand
I guess this just means that Hi stat counts even less than before, with riders like Meurisse getting kind of an "upgrade" instead. Something to get used to, and probably something to consider for the OVL formula - if it turns out this is really the case.
|
| |
|
|
| whitejersey |
Posted on 04-04-2024 23:03
|
Classics Specialist
Posts: 2904
Joined: 07-08-2011
PCM$: 300.00
|
I will have a more constructive comment at one point but all I'll say is that if Meurisse gets another OVL hike and TTers aren't addressed in either OVL or race schedule then it's incredibly uncalled for. Out of the country atm so can't really put up something more concrete but I touched on that issue further up in this thread.
|
| |
|
|
| jandal7 |
Posted on 04-04-2024 23:08
|
World Champion
Posts: 11392
Joined: 17-12-2014
PCM$: 1020.00
|
Fabianski wrote:
I'm not judging whether it's positive or negative - it just feels different from the past version. It's something managers will need to adapt to - this just wasn't really possible without knowing it beforehand
I guess this just means that Hi stat counts even less than before, with riders like Meurisse getting kind of an "upgrade" instead. Something to get used to, and probably something to consider for the OVL formula - if it turns out this is really the case.
Fair enough. Two points here. To the first one - I'd say of all the curveballs we've been sent by PCM (not on behalf of the MG runners but from Cyanide changing things) that "riders with bad sprint aren't good sprinters" is probably one of the ones that managers could very easily have seen coming
To the second point - I agree with your findings about Meurisse-type but not your conclusion (and I hope that now he doesn't ride for me that people can listen to me on him) - he is definitely doing better but is that worthy of him getting an increased OVL, or him finally justifying that the OVL formula always valued his well-roundedness a little too highly and he was in fact underperforming for a long time?
He's got the 44th highest OVL (77.14) in the PCT and as of the last update is 26th and falling in the standings. Yes that means he's overperforming his OVL, but I would say less so than riders like Izagirre (73.55, 27th), Gamper (76.9, 22nd), Powless (77.07, 60th but will end much higher after ToA), Vesely (77.15, 12th), Afewerki (76.07), van Hooydonck (76.96, 18th), Abdul Halil (76.1, 20th), Paillot (77.09, 15th). Masnada (76.1, 35th) and Jorgenson (75.81, 34th) are both subjective, I would say they are bigger "overperformers" but whatever - they are in the same league as Meurisse). Spengler (77.19, 31st), Aniolkowski (77.05, 30th) and Zabel (77.35, 25th) are all having very similar seasons to Meurisse at this point.
You then can go to the likes of Theuns (77.79, 6th) and Manninen (77.68, 10th), who have about half an OVL point higher, around 30-35th in the division, and are scoring far closer to the top riders than their compatriots in OVL.
Izagirre is the only other puncheur I found in this cursory look, and he is actually relying on a big HI stat as well as another main stat, not being rounded like Meurisse.
I'm not sure what can be fixed in the OVL to take this diverse group of riders from overperformers to regular performers (well, I can see one stat that's consistent between a lot of them but nobody wants my thoughts on that...), but it's either a much bigger issue than puncheurs if we want to change the OVL in a big way (which is what would be needed if we decide to do OVL changes in my opinion), or we can decide it's not an issue of OVL. Maybe wages could have even more tie to performance than OVL, so when deciding what is a "fair balance" we look more at who outperforms their wage instead of their OVL? Maybe it's a mismatch of what race types the OVL values and what is most present in the calendar?
I also think there's also a certain element about how riders perform in different divisions that accounts for OVL doesn't match to performance exactly. For example, while the OVL is correct in suggesting Areruya is likely to finish as the best rider in the PCT, I find the idea he'd be finishing 8th in the PT laughable. Going the other way, Saber could often falter in the PCT compared to his OVL, but it was obvious that he would destroy the CT last year and he duly did. Obviously doing anything to account for this with dynamic OVLs, wages, etc. for riders going into different divisions with different roles or chances of doing well is far too complicated a game mechanic to just introduce and for me goes against a lot of the ideas of MG. But just another note on why this happens.
This was a very short "study" that shouldn't be taken as anything other than "it's not just the puncheurs" - I didn't look at what rider types are underperforming, I didn't look at outliers of exceptionally poor race selection or luck, and most importantly the season isn't over. Just wanted to make that point as while you're right Meurisse is overperforming his OVL now (after his OVL used to be far too high compared to points scoring) I don't think it's as widespread in the hills, and riders' performance not being tied at the hip to their OVL is not a new issue, and potentially not a huge one either. |
| |
|
|
| Fabianski |
Posted on 04-04-2024 23:14
|
Grand Tour Specialist
Posts: 4667
Joined: 29-09-2018
PCM$: 185.00
|
whitejersey wrote:
I will have a more constructive comment at one point but all I'll say is that if Meurisse gets another OVL hike and TTers aren't addressed in either OVL or race schedule then it's incredibly uncalled for. Out of the country atm so can't really put up something more concrete but I touched on that issue further up in this thread.
It's not necessarily riders like Meurisse who need a bump, maybe rather pure puncheurs like Turgis (or Boswell) who'd deserve a downgrade... The issue is that OVL formulas would have to be looked at knowing the final rankings, to compare if OVL of different rider types corresponds +/- to similar points - but then it's basically too late to adjust them for the next year ^^
Fully agree on TTers being underrated, though. I do obviously like their schedule (at least the TTTs), but having an 80 TTer at about 100k wage - when TT apparently is the only stat that matters in flat TTs - is pretty weird.
jandal7 wrote:
Maybe wages could have even more tie to performance than OVL, so when deciding what is a "fair balance" we look more at who outperforms their wage instead of their OVL? Maybe it's a mismatch of what race types the OVL values and what is most present in the calendar?
Definitely agree that this sounds more appealing than always trying to adjust OVL to latest evolutions. But how would the tie look like? Points? Wins? A "fair balance" definitely would need to be found.
You named Masnada as an example of an overperformer - which I can't deny obviously, but which is also closely linked to the team strategy of going all-in for TTT-heavy races. Quite some part of a rider's score in general depends on planning, and sometimes on team setup - how to calculate a "fair" wage?
Edited by Fabianski on 04-04-2024 23:25
|
| |
|
|
| jandal7 |
Posted on 04-04-2024 23:55
|
World Champion
Posts: 11392
Joined: 17-12-2014
PCM$: 1020.00
|
Fabianski wrote:
You named Masnada as an example of an overperformer - which I can't deny obviously, but which is also closely linked to the team strategy of going all-in for TTT-heavy races. Quite some part of a rider's score in general depends on planning, and sometimes on team setup - how to calculate a "fair" wage?
You're absolutely right, and, though I think he's slightly undervalued in his own right, his team's TTT strength (and the prevalence of time trials) is a huge reason as to his performance. That kind of factor is the missing nuance that I alluded to.
I'd be happy to go through and look more in depth at factors that cause the discrepancy when the season's done as it's just a subject that interests me as a manager going forward so I might do it anyway. As we both seem to agree there's a variety of fixes, and a debate to be had over whether it needs fixing in the first place and how much  |
| |
|
|
| quadsas |
Posted on 05-04-2024 00:11
|
Small Tour Specialist
Posts: 2518
Joined: 18-01-2013
PCM$: 300.00
|
Fabianski wrote:
Fully agree on TTers being underrated, though. I do obviously like their schedule (at least the TTTs), but having an 80 TTer at about 100k wage - when TT apparently is the only stat that matters in flat TTs - is pretty weird.
I say we postpone that discussion until 2027 at least for no selfish reasons (and I think free agency counteracts that point regardless)
|
| |
|
|
| SotD |
Posted on 05-04-2024 10:26
|
World Champion
Posts: 12186
Joined: 29-11-2006
PCM$: 2980.00
|
OlegTinkov wrote:
In my humble opinion everything works as good as you can expect as long as you provide the proper team/stagegy. Meaning if you sign a top sprinter you at least need to build a team around it with leadouts and second sprint option. Same goes for almost every terrain, you just can't have 1 top rider - fill in every terrain - and expect it to work.
Most teams with top sprinters have a pretty solid leadout - Doesn't make a difference in this game.
|
| |
|
|
| Ezeefreak |
Posted on 11-04-2024 07:19
|
Domestique
Posts: 555
Joined: 06-07-2009
PCM$: 300.00
|
It's me again! 
At this time at the year i seem to have too much free time and if I have too much it seems i think about too much useless stuff. This behaviour lead me now that I have a new suggestion 
I was looking at the training /leveling of not maxed riders and did some analysis. Too long to write up my complete analysis here, so i summarize it quickly:
Training/Stagains seem pretty unbalanced, for example Classics & Cobbles Training have constantly 30 to 45 % less points gain than Fighter for example. I understand this probably results because it just historically grows and its probably not a huuuuuuuuuge problem. But in the end this will lead to that we will have tons of similar riders in the database because everyone will chose the few most effective ways for their riders. So i think it will hurt the varaition of the database in the future.
So I would suggest an overhaul of the training/leveling. I dont wanna flip it over complete and still make it predictable for everyone. I already have worked out something that could be the base for a new one. My idea in general was to split between the actuals "skills" (Mountain, TT, Sprint and so on) and the Energy/physical attributes (stamina, resistance, recovery & fighter) . So then for each lvl up you chose 2 "training focus", one of the actuall skills and one for energy stats.
For example:
"Sprint + One Day" or "Stage + GC" and so on. In my current version there are 8 Skills Trainings and 3 Energy ones. This means for each Step/Level up there are 24 possible combinations, and over the whole process this are over 300000 possible variations. Compared to now 12, this is then "only" about 20000 possible combinations. So in the end this will lead to just more different but still very useful riders.
I know you think it gets more complicated and more work behind the scenes, but i dont think so. Also I did some comparrison between the current System and what I have thought of, it isnt like everything will changes if you do similar focus on training you will get similar results. So if you have a guy building into a Stage GC guy then with both you get really similar result in the end. The difference is more that you would have the portunity to build something else without making the rider "useless"
Okay enough talkinghere. Like I said I worked out already something to be used as base for further discussion.
For the people in charge then, if you're willing to discuss this just write me a message here or on Skype and i can provide you everything i made so far and we can discuss this.
Thanks for reading till here
Edited by Ezeefreak on 11-04-2024 07:29
|
| |
|
|
| Ad Bot |
Posted on 21-11-2024 19:14
|
Bot Agent
Posts: Countless
Joined: 23.11.09
|
|
| IP: None |
|
|
| redordead |
Posted on 11-04-2024 07:38
|
Grand Tour Specialist
Posts: 4881
Joined: 18-10-2017
PCM$: 200.00
|
A stat gains overhaul is certainly something interesting to look into. I'm not sure if it needs to be to the extent that Ezee layed out, but I wouldn't be against it. Having more development options is always good if it forces the managers to strategise more about how they form their teams. The current stat gains, don't give a ton of flexibility unless you start with a rider that has a less typical stat combination.
"I am a cyclist, I may not be the best, but that is what I strive to be. I may never get there, but I will never quit trying." - Tadej Pogačar
|
| |
|
|
| whitejersey |
Posted on 15-04-2024 14:31
|
Classics Specialist
Posts: 2904
Joined: 07-08-2011
PCM$: 300.00
|
Outlining the PCT Calendar:
TT is the most overpowered stat in MG, especially in PCT. One of the biggest issues is that a lot of the stage races that suit puncheurs for GC points are taken over by either an ITT or a TTT. In reality on the entire PCT calendar, including PTHC, there are four hilly stage races that doesn't include any racing against the clock. These are:
Balkans International
Tour de Pologne(Essentially a Climber/Hybrid race if you look at the last few editions)
Tour Down Under(which essentially has a ton of sprinters in the upper GC race if you look at top 10 over the years.)
Euskal Bizkleta
There are 10 stage races that include hilly stages as GC relevant stage that are dominated by the TT instead. (ToJ still pending)
Tour of Ukraine
Lithuania
Slovenia
South Africa
Tour of Japan
Tour of Norway
Post Danmark Rundt
Hong Kong Challenge
Britain
Pais Vasco
Then you have other races where TTTs are dominant factors as well such as:
Vineyards
Romandie
Isle of Man TTT(obviously)
Tour of America, which this year has 4 seperate TT stages of varying character(Which isn't necessarily an issue as a 1 of every once in a while).
Then you add in a race such as Arab Tour which is essentially just a TT GC race with a few sprints in there for good measure.
At this moment in time on the individual rankings the best PCT Puncheurs are:
27th - Robert Stannard(Has benefited from U25 points this season as well.)
32nd - Xandro Meurisse
33rd - Emmanuel Buchmann
Again I recognize that this isn't necessarily a super useful comment in the grander scheme of things but with how few gaps that have been created on hills stages in these races over the course of the season is really does feel like they're TT stage races where the odd puncheur snatches a top 10 spot moreso than actual competitive races where the puncheurs can make a difference on the stages aren't against the clock.
|
| |
|
|
| Fabianski |
Posted on 15-04-2024 15:01
|
Grand Tour Specialist
Posts: 4667
Joined: 29-09-2018
PCM$: 185.00
|
Lithuania and South Africa have a TTT, no ITT. But you're obviously right that TT has a huge impact in this year's PCT calendar.
On the other hand, puncheurs probably have the biggest amount of classics in relation to stage races, with just Hallstatt for climbers and none for TTers (not counting the TTT).
I'm sure that some fixes can (and probably should) be done, and I remember someone from the calendar team saying they're already on it.
I think that one of the issues with puncheurs is that the "pure puncheurs" don't really work anymore with the recent game versions - at least not in stage races. Sprinters now often can keep up - and even among the puncheurs, the importance of sprint stat seems to have clearly increased, with Meurisse being the best example for that. One-dimensional puncheurs seem to struggle more than ever, which is just due to the way the game has evolved.
So I believe it has become far more difficult to design hilly stage races that are actually for puncheurs. I'm not sure which direction to take, but I'm sure Ulrich already has some ideas.
One thing that could definitely be done would be to switch race categories; currently, there's no race like TdU (which got demoted to C1) or Euskal in HC or PTHC if I'm not mistaken, so swapping one of them for one with a TT might be an idea.
And I do also think the TTers need to be addressed anyway. With the new game version and the restriction to +/- 2 daily form, their results have gone from pretty random (I remember Zmorka outside the Top 30 or so last year...) to probably the most predictable ones. And the TT stat alone seems to be decisive, with neither energy nor flat stat playing much of a role (would need to validate if Mo/Hi has an impact on slightly bumpy stages). So a 78 TTer basically can beat any other 78 TTer or even some 79ers on a good day (e.g. Sterbini in Deutschland), so the TT stat should be far more decisive for OVL calculation.
And in the current calendar setting, with the lower variance we now have, I believe the TTers should be higher rated in general. And yes, I know it would hurt me, but I believe it would be fair.
|
| |
|
|
| KaiserAdler |
Posted on 15-04-2024 16:45
|
Domestique
Posts: 570
Joined: 13-07-2022
PCM$: 5800.00
|
There was a discussion about this in a similar small scale project like this about how TT works. And as far as I remember. Up until 15km its pure prologue stat, up 30 its a mix between pro/tt and past that pure tt and flat/energy. But don’t quote me on that.
|
| |
|
|
| TheManxMissile |
Posted on 15-04-2024 17:14
|
Tour de France Champion
Posts: 18187
Joined: 12-05-2012
PCM$: 0.00
|
redordead wrote:
A stat gains overhaul is certainly something interesting to look into. I'm not sure if it needs to be to the extent that Ezee layed out, but I wouldn't be against it. Having more development options is always good if it forces the managers to strategise more about how they form their teams. The current stat gains, don't give a ton of flexibility unless you start with a rider that has a less typical stat combination.
This is the basis of what i've been saying about more rider additions at Lvl2/3/4 and at lower Pot, 1-4. This way you can add riders with more varied stat combos that don't become a risk of being totally OP.
For proof, see basically the entire T3A team. The mixes, the combos, are far more interesting as a manager, and it makes for a more varied DB.
And in turn it helps manage inflation by widening our stat curves, having faster declines. And is a safer route to adding unique nations/regions as it makes them more viable without risking the Sri Lanka problem again.
|
| |
|
|
| ivaneurope |
Posted on 15-04-2024 17:53
|
Classics Specialist
Posts: 2933
Joined: 09-05-2011
PCM$: 300.00
|
I'd rather limit the number of riders added per year - either as a fixed number or proportionate to the number of riders removed from the DB. And for lower Pot riders perhaps their decline should start earlier while for the highest potentials to be pushed 2-3 years.
Riders with Pot 3-5 is the most optimal range for rider additions. Flooding the DB with low quality Pot 1-2 riders wouldn't be wise and would deter rider development of this type of riders which would be domestiques at best
|
| |
|
|
| quadsas |
Posted on 15-04-2024 18:00
|
Small Tour Specialist
Posts: 2518
Joined: 18-01-2013
PCM$: 300.00
|
KaiserAdler wrote:
There was a discussion about this in a similar small scale project like this about how TT works. And as far as I remember. Up until 15km its pure prologue stat, up 30 its a mix between pro/tt and past that pure tt and flat/energy. But don’t quote me on that.
almost certain it's not 15km. it used to be <3km for PRL only and then sliding scale up to x km
|
| |
|
|
| Fabianski |
Posted on 15-04-2024 20:44
|
Grand Tour Specialist
Posts: 4667
Joined: 29-09-2018
PCM$: 185.00
|
quadsas wrote:
KaiserAdler wrote:
There was a discussion about this in a similar small scale project like this about how TT works. And as far as I remember. Up until 15km its pure prologue stat, up 30 its a mix between pro/tt and past that pure tt and flat/energy. But don’t quote me on that.
almost certain it's not 15km. it used to be <3km for PRL only and then sliding scale up to x km
Definitely not 15, no. But likely also more than 3km. Looking at the results suggests that up to 10km it's still mainly PRL stat, with Schomber clearly winning e.g. an 8.4km prologue in California. It isn't purely PRL though, as Kopfauf who can't TT at all just took 23rd, while taking 2nd in a prologue that was 2km shorter.
So I suppose it's 100% PRL up to 5km, then gradually sliding over to TT. Don't know when it's 100% TT though, my feeling says about 20km, could also be 15km.
The flat/energy theory would however be interesting, as I can't really observe that in TTs so far.
|
| |
|
|
| TheManxMissile |
Posted on 15-04-2024 21:19
|
Tour de France Champion
Posts: 18187
Joined: 12-05-2012
PCM$: 0.00
|
ivaneurope wrote:
I'd rather limit the number of riders added per year - either as a fixed number or proportionate to the number of riders removed from the DB. And for lower Pot riders perhaps their decline should start earlier while for the highest potentials to be pushed 2-3 years.
Riders with Pot 3-5 is the most optimal range for rider additions. Flooding the DB with low quality Pot 1-2 riders wouldn't be wise and would deter rider development of this type of riders which would be domestiques at best
I've also been saying this for years. New additions should basically equal riders exiting the DB! Any upgrades matched with equal downgrades. etc. etc.
I like declines being fixed age. It's simple, for both managers to understand and plan around, and for the DB to be updated each season.
Whilst yes Pot 1-3 are not what managers might want, you can still easily make them enticing and usable, for years. I just advocate for a bit more variety, and certainly not a flood of riders for no reason, and as my post explained it makes a great way to build in varierty and variation to riders by having lower potentials paird with higher starting stats
|
| |
|
|
| Ezeefreak |
Posted on 16-04-2024 08:02
|
Domestique
Posts: 555
Joined: 06-07-2009
PCM$: 300.00
|
About the Leveling. I did some more looking into the Stat gains. In the Table below is shown how many Stat Points in total you get for each Level Up at each training (also for each potential). Sure in the end it matters what Attributes are raised but this is just to visualize that pts are distributed not really equal.
I didnt include the sprinter bonus also because its conditional and doesnt apply to every rider.
In my opinion this should be equal or at max 1-3 pts difference OVERALL. But here it is up to 20 pts. So its naturally, even if you didnt dig in so deep most rider will be developed in a similar direction.
|
| |
|
|
| SotD |
Posted on 16-04-2024 09:59
|
World Champion
Posts: 12186
Joined: 29-11-2006
PCM$: 2980.00
|
I still think moving back to the previous version is the best possible solution we could ever pick of the Man-Game. We can tweak and twist all we want, we just won't get as good a result as by moving backwards.
I know it's not what people want, but sometimes there isn't a correlation between what people want, and what people need.
|
| |
|
|
| quadsas |
Posted on 16-04-2024 10:05
|
Small Tour Specialist
Posts: 2518
Joined: 18-01-2013
PCM$: 300.00
|
SotD wrote:
I still think moving back to the previous version is the best possible solution we could ever pick of the Man-Game. We can tweak and twist all we want, we just won't get as good a result as by moving backwards.
I know it's not what people want, but sometimes there isn't a correlation between what people want, and what people need.
should be a perma ban for such ridiculous suggestions
|
| |
|
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
