As said couple of weeks ago, I want to get a discussion started how to improve the game or how to go on once the season swaps. To get a good picture, this thread is supposed to start a discussion. Ideally we are getting a lot of AI knowledge of people, who have experiences with PCM versions and how the MG would work on different versions.
Some things to consider as it`s pretty much needed to keep in mind when talking about a potential version change.
- PCM Stages are sometimes awful to convert and sometimes this is not even possible. We now have a huge selection with PCM18 stages (which to my knowledge are 100% convertable or direct-to-use for PCM19 and PCM20, unfortuantely not backwards to PCM10-17)
- Whenever there is a game change, DB tables might have changed, which can make DB editing in the offseason more complex, e.g. equipment.
- PCM DBs and MG DBs are very different. So having a career running on PCM with good AI does not necessarily mean good AI for the MG.
- Game Settings, e.g. Hill/Mountain Ratio were changeable with older stages, not to 100% with PCM18 and not sure if PCM19/20 changed anything again, probably not as all stages are usable from PCM18.
- Clear AI issues in specific PCM versions that are not ideal for MG use?
Please consider more points as those above, but if you are able to give information to any of those, feel free to start a discussion already.
I am open to a game change as also staying with the game as later would mean, we know what to expect, which then should have known price tags for specific rider types etc. Something that is less known whenever changing a game. So there clearly will be pros and cons for both options.
Going backwards is always a possibility as well, but this is usually a lot more difficult, mainly due to not having the pool of stages, we have used the last 2 seasons and neither PCM15 was ideal e.g. acceleration stat, top sprinters being ignored regularly etc. So I would not be a big fan of going backwards in PCM versions myself really.
- Regarding OVL Formulas. This is looked into anyway! Last year it was adjusted a bit already according the needed changes from PCM15 to PCM18. Smaller adjustments after the experiences of another season are very likely to come in, not as big as last year as this was the bigger update, but adjustments as usual.
- In addition to the still active performance ratio, which also will be adjusted a bit, this should lead to a pretty fair wage demands whenever you have a rider, that on paper is strong but fails with the actual PCM AI. Good examples are the low ACC riders in PCM15, low mo puncheurs last season, low mo timetrialists or sprinters in general due to the way PCM18 handles sprinters.
Edited by roturn on 04-02-2021 13:46
Defo fo sho. Change the OVR formula for mo/hil hybrids. Ayub, Carpenter, etc. Their performance has absolutely no correlation with their salary demands.
quadsas wrote:
Defo fo sho. Change the OVR formula for mo/hil hybrids. Ayub, Carpenter, etc. Their performance has absolutely no correlation with their salary demands.
Carpenter doesn't fit on this list, you literally grabbed him out of free agency.
Added 2 more points regarding the OVL stat. As usual OVL-Performances are looked into and likely get adjusted. Not as big as last year with a bigger update, but here and there improvements clearly can be done.
Obviously the Free Agent wage for riders must be divided from riders on long term contracts here as one is basically coming from formulas, the other from whatever you are willing to spend or overpay from FA.
quadsas wrote:
Defo fo sho. Change the OVR formula for mo/hil hybrids. Ayub, Carpenter, etc. Their performance has absolutely no correlation with their salary demands.
Carpenter doesn't fit on this list, you literally grabbed him out of free agency.
looking at DB most of the renewed hybrids are at least 120k with their overall in high 75s or 76s, while only having MO and HI stats at that level. Someone like Henao is 125k and his overall is 75,67, with 76/76 and no other stat above 73. The formula doesn't make much sense for those type of riders specifically
The results of a rider should be taken into account more than before compared to his OVL. I think this would definitely help managers with riders that let them down time after time and still demand very much. It wouls also partly solve the problem of recently maxed out rider without any result to suddenly demand what the established riders get.
One thing I would note here is that we should check availability among reporters. If only a handful have it, that might be an additional problem.
I could definetely do some testing for PCM20 AI behaviour in the future (likely from March on). Haven't actually played the game a lot myself so far, so I can't comment on that too much yet.
I agree on re-working the Wage calculation to match closer to Points than OVL.
This will go a good way to addressing the related issues for Sprint focused teams, and those with sub-top tier cobbles riders. Will also better react to the specific AI issues year-to-year vs trying to re-make stages.
Granted a downside to this would potentially hit riders turning 25 that are no-longer U25 eligible, but as the theory of the game is that riders don't max until this point the effect should be small/balanced by their reaching max in stats.
Put Cobbles back to 5*
Cobbles are already restricted in days, can we please give those days significance?
My alternative would be to drop CB impact on the OVL calculation or add more CB specific Stages/One-Day races.
Bring back Crashes
This will be wildly unpopular.
But given Punctures cannot be stopped, and neither can random good or bad AI days, i'd like to see crashes brought back.
My concession to this would be that for Stage races 8 days or longer the rate of crashes is dropped, or crashes removed
Update PT WildCard to Include Actual Team Selection
This year it felt really bad, but crops up every year i can go back and look at. I know we want to give people a chance to ride PT races, that's fine.
But lvl4 XP is now gained at PTHC, so when a team only sends unmaxed riders, or worse just doesn't send the best and you can only watch having also applied. It's not a good look for the supposed best races in the game. Quite simply as part of the application teams guarantee X riders will compete and this is factored in.
I also suggest these can become scoring races for PCT Teams to help boost this interest, and to balance WC cost can go up or maybe cost some RD.
Inflation
See previous years and discussions, it's still a problem.
Removing the "Transfer Tax" for CT Teams
Training is still something mostly reserved for PT teams, and i just think this could encourage a few more CT teams to train a little. And with the still low amount would focus training towards role-play reasons or all-rounders.
Also for CT teams, or maybe even PCT, i'd like to see us try and introduce another reason to use money apart from buying riders and training (i know PCT has PT wildcards, but you only do those for specific XP needs or had money left over).
For CT this could be buying an extra Free-Wage rider slot, and PCT teams buying one at all. Or some form of Cap-easing, spending cash to get extra wage cap allowance/buy down a riders wage cost.
Could be limited in all sorts of different ways, but just trying to give ways to make things a bit more varied in transfers and find ways to assist keeping inflation as low as possible.
TheManxMissile wrote: I agree on re-working the Wage calculation to match closer to Points than OVL.
Yes
Put Cobbles back to 5*
Cobbles are already restricted in days, can we please give those days significance?
My alternative would be to drop CB impact on the OVL calculation or add more CB specific Stages/One-Day races.
Look at Andorra, who only have cobblers. People can survive even if they solely focus on cobbles. No point in changing, though I was always in favor of 5*Cobbles. Keep each race unique with the cobbled stars.
Bring back Crashes
Hell no, it's just like more punctures.
Update PT WildCard to Include Actual Team Selection
no, WC spots should decide themselves what they want to do with it
Inflation
Measurements are running already.
Removing the "Transfer Tax" for CT Teams
Doesn't make senseas almost no CT teams will ever reach taxable regions and if they do they have a massive advantage over the other teams anyway.
Also for CT teams, or maybe even PCT, i'd like to see us try and introduce another reason to use money apart from buying riders and training (i know PCT has PT wildcards, but you only do those for specific XP needs or had money left over).
For CT this could be buying an extra Free-Wage rider slot, and PCT teams buying one at all. Or some form of Cap-easing, spending cash to get extra wage cap allowance/buy down a riders wage cost.
I will give my full thoughts when I have time later and the discussion has developed, but I guess my main concern is this:
There are AI problems in PCM 18 which add a degree of randomness into the game that is hard to compensate for. In some areas, like hills, we have been able to remedy those issues to a degree with careful editing. In other areas, as we saw in the Giro, it doesn't seem like much can be done to make teams smarter (Generali). And in sprints specifically, I think the energy system the game uses just fundamentally doesn't work, especially given our DB.
So, I believe the main issue is still sprints. And I think going on like we have been, where 75% of teams are complaining in each thread about every sprint stage, is untenable. I'm one of them more often than I'd like to admit, and I recognize how bad this toxicity is for reporters and in general keeping people motivated and active.
But with the level of randomness and just pure nonsense going on, such as main sprinters just not following trains and trains randomly sweeping across the road to ruin everything for no reason, I have to say that staying as is just isn't viable either. And unfortunately, making the roads wider and straighter only seems to have worsened the problems. And although I've seen others theorize otherwise, I have not seen success in this scenarios actually stratified by a certain stat group like people claim. Flat and resistance have certainly not become king as many people said, so the changes cannot be easily offset by OVL changes like ACC in PCM 15.
So the question becomes: how can this be solved? I'm not sure it can be with the energy system in PCM 18 honestly. And we cannot have main sprinters sitting back and just letting half a dozen other sprinters go up the road in finales for no good reason anymore. As cio mentioned in the Giro thread, it just goes against the system of OVL and wages too much to account for in team building. I know personally how infuriating it is to see AKA not even bother sprinting sometimes despite investing over a million to train him and get him a top leadout man (who hasn't actually done his job once so far).
After two seasons on PCM 18 I doubt many people will be in favor of returning to PCM 15 even with the problems. But what else can we do? Does anyone have any ideas? Upgrading to PCM 19 or 20 is not an option I'd necessarily object to with enough testing, but if the basic energy structure and AI is the same, I'm not sure it will make much difference.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
If some reporters wish to continue, but doesn’t have PCM20 I am interested in donating a version for a reporter. Maybe others feel that way too, so IMO that shouldn’t count as an argument against a change.
I will continue to read on your input, and try to reason my own input some time later, as I don’t know the games myself.
SotD wrote:
If some reporters wish to continue, but doesn’t have PCM20 I am interested in donating a version for a reporter. Maybe others feel that way too, so IMO that shouldn’t count as an argument against a change.
I will continue to read on your input, and try to reason my own input some time later, as I don’t know the games myself.
I second this.
And we might want to make a decision about this before the Steam Summer sale as I think we could get the game much cheaper then especially with a newer version coming out at about the same time.
I'm not sure if those really have any huge differences from PCM18, but a regular DB clearly won't give us the expected results so I guess we don't know what will happen in the MG until we actually try it.
For me going back to PCM15 isn't an option as we'd be stuck with that game for ever and we'd get in a boring loop which would eventually kill the game.
I think the best option is to try to change to a newer game and hope it goes well as well as making small adjustments to the database(when we edit existing riders or add talents) and stage profiles to work better with the newer games.
So first of all, PCM 18. I'm guessing a majority aren't happy with it and its AI. To me there are major issues. Firstly, as BBL says, the sprints are bad, and pretty random. Theres no skill to it, sometimes a sprinter will sprint from his train and launch at 500m to go and win. Other times he will abandon his perfectly set up train to launch his sprint at 1.2km to go and completely die out 300m before the line. Some might say this adds a nice element of randomness, but the logic from it is missing. If I sent a strong leadout team only for my sprinter to do that, I would be pissed.
Next we have the issue of hills/climbs. I think we've fixed a lot of the issues with stage design, there are still some weird things happening, like Reis or Spilak in the Giro f.e
Now the AI of 1week stage races isn't bad, but in GTs it seems to be horrible. A breakaway won way too many times, we saw it in the Giro, Vuelta last year, and from what I remember of the Tour of America, it was the same there as well. Im fine if the break wins 1/3 of the stages, but it felt like they were winning every single stage. I brought Shikai to the ToA and I think he ended up winning 1 stage iirc, as the best climber in the race. In this Giro he was 6th in the GC, and while I dont think he should be winnning stages, I think he got very minimal points from stage results, which is not what I would have expected from a Top 10 finisher in a GT.
What about TTs? Our db is set up so that most TT specialists are terrible at climbing. With the new AI, they are quite poor if there is an incline of 2.5% or more for longer than about 100m.
Moving to PCM 19: I've heard it handles duo leadership a lot better. The issues are that it still has all the same AI issues as 18, and I read somewhere of someone who did some testing, that Stamina had no affect. I believe I remember reading somewhere that the sprints were different to 18, but I cant remember if they were better or worse, but maybe someone else will have experience or knowledge.
PCM20: The AI is more aggressive. This sounds nice, but it ends up being riders attack from 50km out and die with 5km to go, so those that havent attacked just ride tempo and put 3mins into those that did attack. (It does work very well for cobble classics but not for mountain stages) I've also seen that on flat stages, the GC riders pull for the sprinters, and mountain stages, sprinters pull for the climbers (rather than reserve energy) Stamina also has no affect apparently in 20. Also, we had long range attacks in PCM 15 as well.
PCM 15: we know how it works, and were more or less satisfied with the AI of the game. If we have lost reporters that own the game, it is probably fairly cheap now, being 5 years old. We also likely have the old OVL formulas that we used for PCM15, so figuring out a new one wouldn't take long for upcoming seasons.
I am personally for moving back to 15. I may be in the minority, but the AI vs AI gameplay is much better there than in 18. Playing PCM18 personally, it always feels like its all the AI teams vs the human player. In general, managers are probably expecting to move forwards in game versions, rather than backwards. So if it does come down to moving back to 15, I think either a 1 season ahead of time warning could be fair, but also a vote, so if for example the majority did want to move back to 15 right away, that would also be an option. I have been trying to build my team to suit PCM 18, as Im sure most other teams have as well, so I can understand if others dont want to move to 15 but if we can get past a few of the set up issues of moving back to an older game, I think the season/reporting will be a lot more enjoyable for all, and rewarding for the teams that have the best teams. (Also I know I brought up Shikai earlier, but dont get me wrong, I would be a relegation candidate on either version of the game)
With all that said, I am open to testing of PCM 19 or 20, but from what I have seen, they are either inferior games (for our purposes) or possibly unusable, with Stamina stat that has no affect, but we have changed so much to the DB or OVL system or stages to suit PCM 18 better, that it might be extremely annoying to move back to 15 at this point.
To suggest moving back to PCM15 is absolutely mental. There's plenty of time to test out new editions, and we should probably be given converted DB to test things out ourselves so we have an understanding of how game with MGDB works.
If someone would test it would be greatly appreciated to be honest.
Much like the above, PCM 15 also works the best for me - but like others suggest I do believe it is the wrong path. Moving back would be an option if this version was completely mindende, and while it is in quite some races I don’t see the game so broken that we cannot keep on working our way forward.
I haven’t particularly built my team for either version tbh. I have signed the best french and greek riders I could, and adjusted slightly to my knowledge and experience of what works well and what does not. F.e. signing Zoidl and Nizzolo was done because of last season, but both have pulled off disastrous results throughout. So in all fairness I haven’t really found an obvious path.
Cobbles and Hills seem to work much better this season - still flawed, but PCM 15 was too.
The mountains seem very similar only with actual High MO (and FIG?) outgunning High ACC. Imo it is better to have 83-85 MO riders be the best, rather than 81 MO high ACC riders regularly beating 83-84 riders.
The sprint issue for me is massive, but I reckon it’s mainly because of the heavy investment put into Coquard.
Like mentioned before a way of making top sprinters up their worth in this randomness could be to allow them to join a GT and spend just 10-12 RD rather than 21, it would also make for some more realistic scenarios I would think with a couple of hardhitters working to keep things tight rather than accepting the BotD taking home the gold all the time. A bunch of 79-81 sprinters just doesn’t suffice. Maybe adding 2-3-4 good sprinters to that list could see some effevtive testing? How would this Giro have panned out with Degenkolb, Bewley and Coquard there with their trains? Maybe test the same with puncheurs of a certain caliber to see if they would keep things close atleast in the frist 7-10 days?
Imo the problem could be that the BotD is often involving riders almost as good as the main helpers in the bunch.
Given that I have no info on PCM 19-20 I honestly don’t know what I want, but regardless of my love to PCM 15 I wouldn’t want to og back again.
I am one of the few that haven't really had any issues with with 18, I however had a ton of issues with how 15 handled my teams in the past (hills armada and that).
A move back to 15 could potentially mean that we just go back to mathing everything which seems to be the wish of multiple PT managers when you look at discussion threads.
The stages themselves are much more interesting to follow than they were in 15 IMO, outside of the last World Champs where the game showed that it could actually be interesting. IMO the play will always be to move forwards not backwards.
Sadly we will never be in a position to carpet bomb the DB to fully rebalance but I do think that the current DB is absolutely mental and will most likely break any version of the game due to how power crept it is.
Overall there's 55 riders with more than 80 MO in the db, 22 of them are in the PCT. These are riders that in more or less every DB would make it into the 2 star favourite field in any given stage when you throw them all together I can see why the AI can struggle. Furthermore I think it's a bit extreme that in the sprint field alone you'll often have to fetch yourself 80+ sprinters to have a shot at CT glory. I am fully aware that it isn't that straightfoward, especially with PCM AI but when you have 7 riders with 80 or more in a main stat in CT it starts to become quite clear how power crept the DB actually is.
I do think that giving some training options to lower ranked teams could be interesting but I have a hard time envisioning how it can ever become reality when inflation is still an issue.
SotD wrote:
Like mentioned before a way of making top sprinters up their worth in this randomness could be to allow them to join a GT and spend just 10-12 RD rather than 21, it would also make for some more realistic scenarios I would think with a couple of hardhitters working to keep things tight rather than accepting the BotD taking home the gold all the time. A bunch of 79-81 sprinters just doesn’t suffice. Maybe adding 2-3-4 good sprinters to that list could see some effevtive testing? How would this Giro have panned out with Degenkolb, Bewley and Coquard there with their trains? Maybe test the same with puncheurs of a certain caliber to see if they would keep things close atleast in the frist 7-10 days?
I agree messing with sprinter race days would be an interesting remedy value wise, and it's one I hadn't thought of. However, as a CT manager, I think the GT idea would only address sprints in PT. This kind of stuff isn't just at top levels of the DB, it necessarily trickles down too. A lot more would need to be thought out in that regard, but I think the race day change idea could be good. Only problem would more more top sprinters ending up at the same races.
On another note, WJ's comment got me thinking a lot. We've been talking about inflation for years now and it clearly does contribute a lot to these issues. I talked in the previous thread last year about how bad it is now. While it has been getting better at the very top levels, it is arguably worse than ever in the 76-79 range, where riders with 76 future main stats don't even get stagiare rides. Too many talents are being added that fall in this range. On paper there will still be differentiation, but in game the effects will be negligible and random once daily form is in play. But that's a discussion for another day.
Anyway, I think perhaps a big issue with the inflation has been exacerbated by this strength of daily form in these newer PCM version. The differences between riders mean little because on any given day 83 and 78 climbers might swap stats or even worse from a +3, when it used to be much less of a factor in my experience. If there was some way we could lessen the impact and spread of daily form effects without eliminating them entirely, that could be a huge improvement. Probably not possible though.
A more realistic thing I can think of would be to have more stages races with terrain variation be common. Different teams might then prioritize different rider types for the same races, which makes the startlists more varied and the gaps in skill level between favorites bigger. So instead of stages races catering to one terrain, and thus being stacked with every rider in the division specializing in that terrain (meaning a spread of like 75-83 in the same main stats for the whole peloton), different teams will choose different races for their leaders. Clashes could maybe help this too?
There are big downsides to that though too.
1.) It would not work with GC points as they are currently constructed - mountain riders would still win every race and thus benefit the most, so it might just make stage racers, who are already the most valuable and have the most powerful XP path, even stronger.
2.) We have seen even in diverse start-lists that the AI is not necessarily better. E.g. The random sprints did not go away in Tachira even with big talent gaps. Rather it was even more absurd because 72-73 sprinters were beating 78-79 ones.
I hold though, that keeping everything as it is with the way sprints currently function cannot really be an option. They are fundamentally broken, and it destroys the market for an entire leader type in a way that is untenable.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
One quick note: if the current state of the DB does not suit the needs of the latest versions of PCM - why don't we just mass downgrade everyone like -5? If 80 becomes new 85, it may not change anything really, but it may also allow the AI to behave differently. I remember how PCM AI behaved differently all the years back in cases like where I pushed all the riders' FL attribute all the way up to 85 - the AI was then behaving differently. Things like this shoud be tested.
Put Cobbles back to 5*
Cobbles are already restricted in days, can we please give those days significance?
My alternative would be to drop CB impact on the OVL calculation or add more CB specific Stages/One-Day races.
Forgot to touch on this yesterday. Having been a part of the testing crew of the DB when the initial switch to PCM 18 was made, it became quite apparent that having cobbles stages in stage races tends to mess things up. The main example of this was using the 2018 variant of Tour de France, where the GC was decided by which GC favourite had the highest cobbles stat rather than in the actual mountains. I do think that it would nice to see more races like Cheshire and BeNeLux Challenge, that offer cobbles riders more chances to excel.
In terms of Roman's suggestion, I don't think that is feasible since you end up removing a ton of teams' leaders in one swoop. For instance Faiers and Diaz has this year done somewhat ok things at 77 and 78 MO, they'd be absolutely useless at 72-73. So whilst it might tackle the actual issues at hand it creates a seperate issue in the other end of the DB which then would need a fix and then we can start a new cycle of balance changes.
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
