|
Weird change in projected rider fitness throughout the season.
|
| ShortsNL |
Posted on 18-03-2014 23:55
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Okay, I present you two screenshots:
This one is from January 8th, when I just planned my rider's fitness schedule.
This one is from the end of March, a couple of months in. Notice the bars stayed unchanged, yet the projected fitness has dropped. Freshness is Excellent, and No Tiredness.
I ask you, what the **** is all this? How the heck is the projected fitness dropping? I'm pretty pissed off since I spent 4 weeks planning my season and fitness and now my game is ****ing me over.
Edited by ShortsNL on 18-03-2014 23:55
|
| |
|
|
| atlanta |
Posted on 19-03-2014 00:17
|
Protected Rider
Posts: 1220
Joined: 31-07-2011
PCM$: 200.00
|
Hi Shorts i cant see the screens, but in past i have had problems whic i think is different than you.
People say the race days dont affect nothing but im not sure, i once had my rider do the Giro/TDF double and in the Giro he did not have the highest bar(saving 2 for the TDF) but beucase i raced around 25 days b4 the Giro after the Giro the TDF bars were showing he would be really tyred. Not sure this is similar to yours just saying. |
| |
|
|
| ShortsNL |
Posted on 19-03-2014 09:30
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Surely atlanta and me can't be the only ones who experienced this. I'm only 3 months into my first ever PCM13 season after all.
Is there anyone who's got any more experience on this? Is the 'no race day limit' thing really a myth then?
|
| |
|
|
| murfi |
Posted on 19-03-2014 20:20
|
Junior Rider
Posts: 31
Joined: 19-08-2013
PCM$: 200.00
|
Cant see the images you posted either but I have noticed this all the time too... I spend ages setting the objectives and fitness curves for the key races and trying to limit the "red" area towards the end of the season... then after a few months, there's a huge red area on the graph for the Vuelta, Lombardy etc. I can barely find enough riders to take part in Beijing!
Personally, I don't think this is associated with race days as much as by number of objectives... for example I have a good regen in my game who is great for the Spring Classics and his targets are usually MSR, Cobbles and Ardennes, followed by downtime until the Vuelta. So even though he will only have 20/25 race days or so by the end of April, his fitness curve is already full of red for the Vuelta. Meanwhile my GC guys targeting the Giro/ Tour who are doing the likes of TA, PN, PV, Catalunya etc are not as badly affected (at least until after the first 2 grand tours)
Anyway, it may be worth looking for a correlation between tiredness and the number of early season objectives...  |
| |
|
|
| samdiatmh |
Posted on 19-03-2014 22:45
|
Breakaway Specialist
Posts: 892
Joined: 29-12-2009
PCM$: 200.00
|
murfi wrote:
for example I have a good regen in my game who is great for the Spring Classics and his targets are usually MSR, Cobbles and Ardennes, followed by downtime until the Vuelta. So even though he will only have 20/25 race days or so by the end of April, his fitness curve is already full of red for the Vuelta.
that's what he has here
his guy is set up to be racing:
ParisNice, MSR, AlPaisVasco, AmstelGold, LBL - before literally nothing and then the Worlds and Lombardia
it should be okay, but you're probably over-racing him
also, if you right click and press "view image" then you'll be able to see them
Edited by samdiatmh on 19-03-2014 22:45
|
| |
|
|
| Ad Bot |
Posted on 23-11-2024 02:23
|
Bot Agent
Posts: Countless
Joined: 23.11.09
|
|
| IP: None |
|
|
| ShortsNL |
Posted on 19-03-2014 23:27
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Okay, I've got some more screenshots from several of my riders, to increase the sample on display. Also, different image hoster so I hope you guys can see the images now.
Bakelants, who had a full programme from the season start up to LBL, then a build up to Worlds and Lombardia.
Spoiler original
after 3 months
Joel Bakker, who serves as a reserve rider. He does have a fitness programme, but entered no races so far and only has 16 race days scheduled this entire season.
Spoiler original
after 3 months
Thomas Bonnin, a helper who has the second longest programme this season of all my riders, totaling 115 race days. However, his schedule only has 4 objectives.
Spoiler original
after 3 months
Lars Boom. Different schedule than the previous riders. 6 objectives.
Spoiler original
after 3 months
Tom Dumoulin, with yet another schedule, with 110 planned race days.
Spoiler original
after 3 months
Paul Michon, another reserve rider with only 8 planned race days.
Spoiler original
after 3 months
I'd like to analyze these but I'm going to do it tomorrow, bedtime for me now.
|
| |
|
|
| Lachi |
Posted on 19-03-2014 23:53
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
The prediction is not perfect. Try not to go too hight while planning.
But you can react to it during the season by lowering some bars. |
| |
|
|
| ShortsNL |
Posted on 20-03-2014 09:12
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Lachi wrote:
The prediction is not perfect. Try not to go too hight while planning.
But you can react to it during the season by lowering some bars.
I see where you coming from, and theoretically, it would definitely make sense that there can be a 'random' deviation from the planning. However I don't support this theory, for 2 reasons:
-If it were just a random deviation, we would not only see negative changes throughout the season. Right now, if the fitness prediction changes, it always goes down. Sometimes it stays the same. This means that it can't be a random factor and that there has to be some sort of mechanic that causes this.
-Cyanide doesn't exactly have a good reputation when it comes to these things. If it appears broken, it's probably not because Cyanide wanted it to happen like this 
---
So far I think what we can conclude from these screenshots is:
-It's not caused by too many race days (see reserve riders)
-It's unlikely that it's caused by too many objectives (see Joel Bakker, only 3 secondary objectives)
I think Murfi might be on to something. So far, the riders that seem the least affected all have a less intensive programme very early in the season. I think there is a link between the two: the longer, earlier and more intensive your rider peaks, will cause a bigger drop in projected fitness later on. Why? I don't know, perhaps a big early peak screws up the prediction system somehow.
This afternoon I'll get some more screenshots of my riders. Until then I'd love to hear your thoughts about this.
|
| |
|
|
| Lachi |
Posted on 20-03-2014 10:23
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
If I remember correctly, in previous versions of the game the fatigue grew after a rider had won a race. Maybe this is still happening? |
| |
|
|
| ShortsNL |
Posted on 20-03-2014 10:36
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Unlikely, since the unexpected increase in projected fatigue is also happening already for my reserve riders, who haven't raced yet and will hardly race (and win) at all -> see the screens of Paul Michon and Joel Bakker.
|
| |
|
|
| Lachi |
Posted on 20-03-2014 13:26
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
You are right, that's not very likely.
Do you send your riders to pre-season training camps? The harder they train, the more does the fatigue raise.
The problem seems to mainly affect riders, who start the season early. So it might be related to the fitness a rider has at the start of the season. Looking at the fitness levels posted here some riders might have to train very hard to reach the projected fitness.
Example: Bakelants and Bakker are starting the year at training level 2 and after two weeks of training they should reach a fitness of 64. But if their fitness is not around 50 at the beginning of the season, they have to work very hard to reach that. This could result in higher fatigue.
This could be verified with a rider, who is injured. After his injury he will start training again based on the fitness schedule set at the start of the season. So if you have to riders with the same schedule, you can compare the raise of the fatigue.
|
| |
|
|
| ShortsNL |
Posted on 20-03-2014 16:07
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
I did not use any training camps whatsoever. Your second theory is very interesting though. I am definitely going to test it further tonight. It sounds plausible, and I did some straightforward analysis, using Tom Dumoulin's schedule, to test it.
The question I'm tying to answer is: if a rider starts building his fitness at the start of the season, does he have to work extra hard to get in shape according to the scheduled fitness bars?
Here's what I did:
Dumoulin's bar shows a moment where his fitness stabilizes, when he is still at training level 1. This must be the maximum fitness a rider can get while training at level 1. I simply checked the size of the graph in pixels, and checked the height of the fitness in the graph when it's stable.
It turns out it's exactly 60%. 60% of 80 training fitness (in correlation with your linked post) translates to 48 training fitness. This also happens to be the minimum of fitness level 1 (6 in-game, but I will call it 1 here). It makes sense: at minimum training you can only reach minimum fitness.
As your rider trains at higher levels your fitness increases, until it reaches fitness level 2 (5 in-game). Level should be displayed at 56 fitness or higher, according to the earlier data.
The crucial thing here is that the riders indeed start at fitness that can be below the level 1 minimum of 48, meaning that any increase determined by an immediate consequtive buildup is extra steep, and that this may indeed cause additional fatigue. It doesn't prove your theory, but it does support it very well.
According to Dumoulin's schedule, he reaches 48 fitness after 4 periods of level 1 training. I guess this timeframe depends on the starting fitness.
One additional note that is interesting is that Bakelants is displayed as if he started with level 2 fitness (minimum 56), but when I measure his graph, he is actually at 34 fitness, below the minimum, offering an explaination also for his increased fatigue.
It should be said that I could have simply done this by looking at rider fitness from a January 1 save, but I am at school right now, and I probably don't have a January 1 save from that game anyway.
What do you think of this theory? Do you think we can trust the accuracy of the graph, in terms of displaying the actual fitness in pixel height? They seem to match.
Edited by ShortsNL on 20-03-2014 16:12
|
| |
|
|
| Lachi |
Posted on 20-03-2014 20:09
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
Yes, I think the pixel height corresponds to the fitness. |
| |
|
|
| Shonak |
Posted on 21-03-2014 09:11
|
Tour de France Champion
Posts: 15615
Joined: 16-07-2013
PCM$: 350.00
|
I think it has some to do with the early season shape, that riders are more tired the earlier the higher they peak.
Another guess, and that's just really a guess, is that it depends on the recovery stats. I imagine that Cyanide can't quite develop a real projection, in which the Recovery stat is already included. Instead, it's a rough projection. After each month or maybe only after a certain period, the Recovery stat is taken into account and thus influences the tiredness, in a negative way though.
"It’s a little bit scary when Contador attacks." - Tommy V
|
| |
|
|
| ShortsNL |
Posted on 21-03-2014 10:35
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Thanks for the replies guys.
On the subject of the recovery stat, as far as I know it supposedly only affects Freshness. I must say i am not 100% sure on that. However, this is not a key issue: because in my case my riders are displayed to have maximum Freshness and no Tiredness/Fatigue yet. There could theoretically be a hidden value that is affected by REC, but still the issue also occurs with riders who didn't race at all yet, so REC can't have been a factor there.
There is one more thing that concerns me though: I took a look at Paul Michon's curve. According to the my measurements and the scale, he reached 48 fitness before he started to build form. This is because his starting form was high. Yet he too shows an unexpected increase in fatigue at season end. Remember: he had 0 race days so far, hardly any objectives, and a supposed base fitness of 48 already when he started building form.
This definitely calls for a bigger sample set. I am going to do a test season up to the end of March with testing schedules, to see more cases. I'll post them later this weekend.
Until then, if you guys have any comments or ideas (especially on the Paui Michon vs. Tom Dumoulin comparison and the 48 fitness theory), please let me know.
Edited by ShortsNL on 21-03-2014 10:36
|
| |
|
|
| Lachi |
Posted on 21-03-2014 19:42
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
The graphs are hard to analyse because the fatigue is not shown until it has an effect on the fitness. If you want to dig into this, you might have to extract the data into excel and do some charts yourself.
I think there could be 2 factors:
Initial fitness: According to our theories, the fatigue will grow faster if the fitness is low at the start of the season. This can be tested by applying the same fitness schedule to riders with different fitness. Prediction: After two months, the fatigue is higher at riders whith lower intitial fitness.
Random: The game could have an randomizer so that the fitness is not predictable. This can be tested by applying the same fitness schedule to riders with same fitness. But how can we tell, that it is random and not caused by other logic? |
| |
|
|
| YvesStevens |
Posted on 21-03-2014 19:44
|
Protected Rider
Posts: 1147
Joined: 29-07-2011
PCM$: 300.00
|
Is it really that bad if such red bar appears? |
| |
|
|
| ShortsNL |
Posted on 29-03-2014 23:45
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Ok, time for a quick update on this.
I did a quick and easy fix by putting fatigue back to 0 (or 1, actually). After I did so, I loaded up the objectives again and they look almost completely like at the season start. I guess I could do this because I was still in March and so fatigue didn't actually affect current fitness yet.
The value I changed is value_i_fat_phy and it can be found under DYN_Cyclist_fitness, which in turn can be accessed and modified easily through DYN_Cyclist using Lachi's editor. For the next season, I am going to make sure I make all riders in the DB start with a base fitness of 48, to not have extra fatigue occur.
I looked in my team and I have two riders with no unexpected fatigue increase, who are currently at fitness level 1 and pretty much have been since the season start, at what must be 48 fitness. One has a Value_i_NPT and Value_i_fit of -27452. The other one has a Value_i_NPT and Value_i_fit of -27435. If I'm not mistaken this is how fitness levels are recorded now and we didn't have the exact formula for it yet, but these will be my starting fitness numbers for all riders in my DB next season.
I didn't bother doing much more testing. as after reviewing more of my riders projected fatigue increase, I feel like we can state this theory with reasonable confidence: there is a direct relationship between sub-minimal starting fitness (in conjunction with early peaks) and extra fatigue.
To answer Yves's question: I hope that this helps everyone who takes their season planning as serious as mine. If you don't bother much with season and fitness planning this thread is trivial. But when you are a hardcore planner like me who wants to get the most out of their riders, this issue can be quite severe.
Of course, one huge final thank you goes out to you Lachi for helping me with this issue and sending me on the right track.
Edited by ShortsNL on 29-03-2014 23:48
|
| |
|
|
| Lachi |
Posted on 30-03-2014 14:40
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
You're welcome. Thank you for investigating it. |
| |
|
|
| Ian Butler |
Posted on 30-03-2014 14:44
|
Tour de France Champion
Posts: 21854
Joined: 01-05-2012
PCM$: 400.00
|
Real great of you to have investigated this. Maybe you could write it up shortly as a Tip & Hint for that section on the site? |
| |
|
