|
Post by Genomico on Dec 24, 2013 1:33:26 GMT 1
The difference between Vijver and Seberla's riders on TT isn't very big. About one point more than the skill difference. And about the attacking part, I lost a lot of good results because my rider only attacked 3 times were others did 5. That way you can always think of something why it didn't work. In this case, you would say that the TT difference did it. In that case, how do you declare than the U21 race? I don't think there were a lot of helpers today because ther aren't much leadership points. And if the TT counts as hard as the terrain skill, Gommers has 9. 5 for flat and 4 on TT. I wonder what helper in the peloton had 9 altogether on flat + TT. The reason that Rats and PAN dominate the flat races is that they have cyclist with a better TT skill than others. So yes, many results can be explained purely based on TT skill. It is also the reason that Te Winkel won Sunday's race. You can discuss if that is a right thing or not, but at least it's a thing we already could have tested for 2 seasons, so it shouldn't be a big surprise. In the U21 there were more things happening than just counting skills.
|
|
|
Post by kurtinsc on Dec 24, 2013 5:12:50 GMT 1
Not true I think (but Nike would have to confirm). I believe helpers try to help... but if they can't keep the rider who falls/flats in the group they drop back and help him chase. It doesn't SAY... "rider X tries to help rider Y after his fall but can't get him back"... but the fact he's in the chasing group with him indicates that's exactly what happened. I always thought helpers just made it more likely that a rider was able to stay with the group, not a certainty. I certainly HOPE it's not a certainty... otherwise low-skill helpers would be way too helpful. What I believe is wrong about the scenario is that the helper SHOULDN'T have tried to help the no-tactic rider at all according to the rules put forward. Helpers are supposed to ignore no-tactic, save fitness and other helpers. They are ONLY supposed to help attackers, GC and final sprinters. There is one other possibility (though I didn't think the engine worked this way). If the assumption for helping someone with a flat is that you "give up your wheel", then your helper may have already been dropped when he helped the first rider and thus unable to help the second (main) rider when he crashed. Kurtinsc, that can't be. The main rider fell first. He couldn't be helped. Then, after that, the no tactic rider had a flat. He has been helped, so the helper was still in the peloton, otherwise he couldn't have helped. The main man could not have dropped before since he has flat 5 and was riding on 100% while I had 2 other riders, one with flat 2 and one with flat 3 riding on 70%. So my main man always should have been in the peloton and should have been helped by my helper who also was still in the peloton. And if a flat 2 and a flat 3 rider on 70% are able to follow, it should not be a problem for a helper, flat 3+ and a flat 5 guy to follow the pace after a fall. Well, the timing proves that the "giving up the wheel" didn't happen. I'm not suprised by that though. While I haven't seen the code for the sim, I've always assumed that the effects within each part happen before the results of the part are applied... not "as they happened". That's why I said I didn't think the "give up the wheel" scenario was the way the sim worked. My guess (and again, I don't see the code so it's a pure guess based on how I would code things if I were on the other side and how the rules have been explained): Beginning of the part: All three are together. It is determined that in this section, rider 1 falls and rider 2 flats. Helper is present and tries to help both. Due to the fall having a bigger impact then a flat, the help is not enough to keep rider 1 in the peloton, but is enough to keep rider 2 in the peloton. Part ends. Because rider 1 is an attacker and rider 2 is a no-tactic, helper drops back to help rider 1. The problem... is that the helper shouldn't have TRIED to help rider 2 based on what has been communicated. Again, I'm basing this on the assumption that when a helper is not capable of helping a rider get back after a flat or fall, there is no message telling us of this failure... the fact the rider lost contact is what tells us that. I'm not assuming every failure for a fallen rider to keep up with the group isn't a failure in the simulation, but rather a failure of the helper. I've had many cases where protected riders were not kept in the group with a fall when a helper is present. I've always just assumed my helpers weren't good enough (or perhaps my rider who fell got dinged up a little and his performance suffered). Hopefully Nike will share a bit with us to clear it up.
|
|
|
Post by Ab Normaal on Dec 24, 2013 10:10:14 GMT 1
Not correct. I had only one helper. If he did stay with Gommers how could he helped Negri then? Like I said "Only strange thing is that when Arbuatti helps Gommers, he doesn't stick with him immediatly after the fall (but I guess what happens is that the fall for example costs 2 minutes and one low-skilled helper is making that 1 minute), but the simulation still sees Arbuatti in the peloton. But I don't think that can be called a bug." So what I mean: Arbuatti first helps Gommers in a mathecally way; which means e.g. that Gommers drops 1 minute instead of 2 minutes. But Arbuatti doesn't drop back yet. He does that AFTER the part ended, so he is still able to help Negri as well. So that is the first function for a helper; to help cyclists with a fall/flat. When the part ended the second function of a helper started; dropping back after the best GC/attacker/sprinter dropped behind. And that is what he did. At the end of the part, he was already behind together with Gommers. I respect your opinion and I am very grateful that you even take the time to read it all and try to explain it to me, but I don't believe this explanation. If the TT is that important, Gommers should be able to come back to the peloton with one of the highest TT skills in the race (single attackers can stay away in front of a peloton) and it isn't said in the race report that Arbuatti helped Gommers. If he did, he could never helped Negri. And if the engine still sees him in the Peloton and that he helped Gommers a little bit, but not enough is just a little too much if's and but's.
|
|
|
Post by Ab Normaal on Dec 24, 2013 10:20:50 GMT 1
The reason that Rats and PAN dominate the flat races is that they have cyclist with a better TT skill than others. So yes, many results can be explained purely based on TT skill. It is also the reason that Te Winkel won Sunday's race. You can discuss if that is a right thing or not, but at least it's a thing we already could have tested for 2 seasons, so it shouldn't be a big surprise. In the U21 there were more things happening than just counting skills. Well, it was said that TT was going to have more impact on the mountains, so that wasn't a big surprise, but it also has been said that for the flat the influence wasn't that high because they would paddle along. But that doesn't seem correct. Then it is really easy then for all teams. Just train TT. That just realizes me that this game isn't that varied as I thought it was.
|
|
|
Post by maxz on Dec 24, 2013 15:03:48 GMT 1
Mate, it's not all about TT... Come on...
|
|
|
Post by Genomico on Dec 25, 2013 2:09:42 GMT 1
... Then it is really easy then for all teams. Just train TT. That just realizes me that this game isn't that varied as I thought it was. The game is very varied. In the first 5 seasons noone trained on TT, because it was quite useless. In the new engine it began to be important and everyone had to catch up. If you pull new cyclists it's not that they only need TT. We all have to train TT because we didn't do so so far, not because a cyclist only needs TT. First mountain/flat were most important, then people had to catch up on sprint and now people have to catch up on TT. Seems quite varied for me. And this season in division 1 has shown that also hill specialized (as main or additional skill) cyclists are winning many points.
|
|
|
Post by kurtinsc on Dec 25, 2013 5:14:42 GMT 1
Like I said "Only strange thing is that when Arbuatti helps Gommers, he doesn't stick with him immediatly after the fall (but I guess what happens is that the fall for example costs 2 minutes and one low-skilled helper is making that 1 minute), but the simulation still sees Arbuatti in the peloton. But I don't think that can be called a bug." So what I mean: Arbuatti first helps Gommers in a mathecally way; which means e.g. that Gommers drops 1 minute instead of 2 minutes. But Arbuatti doesn't drop back yet. He does that AFTER the part ended, so he is still able to help Negri as well. So that is the first function for a helper; to help cyclists with a fall/flat. When the part ended the second function of a helper started; dropping back after the best GC/attacker/sprinter dropped behind. And that is what he did. At the end of the part, he was already behind together with Gommers. I respect your opinion and I am very grateful that you even take the time to read it all and try to explain it to me, but I don't believe this explanation. If the TT is that important, Gommers should be able to come back to the peloton with one of the highest TT skills in the race (single attackers can stay away in front of a peloton) and it isn't said in the race report that Arbuatti helped Gommers. If he did, he could never helped Negri. And if the engine still sees him in the Peloton and that he helped Gommers a little bit, but not enough is just a little too much if's and but's. I guess the point that I'm failing to make is that I often look at games like this as programming exercises. It's possible that the code is done in a way so every individual action happens at a certain time in the race, and the impact of that action should make sense chronologically. But to be honest, that would not be easy to do I think. How I believe it works is that every "section" is simmed for each group. The whole of the tactics set for the riders in that group entering the section are calculated in a chunk... the time they occur within the section for the race report is essentially "dressing". It doesn't really have any impact on the end result. It doesn't matter which rider attacks first in the section based on the report... just that they attacked. So essentially you had a rider as a helper when the group hit the section in question. His "helper factor" applied to all the riders in that section who fell or flatted (again... this is wrong. he should NOT have helped the no tactic rider based on the information we were given when the GC option was implemented). His "helper factor" was enough to help the rider who got a flat keep contact. It was not enough to help the faller keep contact. I understand the faller may have had better stats, but this could be because crashes are harder to help back... or it could be that he had a minor injury from the crash that impacted his ability for the race. I don't know enough about the sim to know if this sort of thing is in the code or not... Nike would have to clarify that. But when the section for that group ended, the attacker was behind and the no tactic was with the peloton... so the helper goes with the attacker. This doesn't mean he couldn't have stayed with the peloton if he were no tactic... he could have. He's supposed to drop back and help the attacker. If I'm wrong about how it's coded... well I'd like to see how they pulled it off without having each section as a lump simulation unit. It would seem to be very difficult to determine when a helper loses contact in a section for example and how to apply thier "helper factor" based on when they were dropped. I've always assumed it's more simple... were they in the group when the group hit the section. If yes... they help that section. If not, they don't.
|
|
|
Post by Ab Normaal on Dec 25, 2013 13:16:18 GMT 1
Mate, it's not all about TT... Come on... Not? Well, check today's result then. Of course you need terrain skills. But we all have riders with the needed terrain skills. But the ones with TT skills win. Every single race. You can better have a little less terrain skill and a little more TT skill. And since all our riders have sufficient terrain skill, all you need to do now is train on TT skill. That is also what I meant with not varied. It was, and will be untill all riders have reached 10 on TT. Anyway, perhaps it is only me, but I am losing intetest very fast.
|
|
|
Post by maxz on Dec 25, 2013 22:28:06 GMT 1
Mate, it's not all about TT... Come on... Not? Well, check today's result then. Of course you need terrain skills. But we all have riders with the needed terrain skills. But the ones with TT skills win. Every single race. You can better have a little less terrain skill and a little more TT skill. And since all our riders have sufficient terrain skill, all you need to do now is train on TT skill. That is also what I meant with not varied. It was, and will be untill all riders have reached 10 on TT. Anyway, perhaps it is only me, but I am losing intetest very fast. i think it was much more boring in the past when only FA was important, only one ability and one skill to train. Just my 2 cents.
|
|
|
Post by NikeBoy on Dec 26, 2013 12:51:42 GMT 1
How I believe it works is that every "section" is simmed for each group. The whole of the tactics set for the riders in that group entering the section are calculated in a chunk... the time they occur within the section for the race report is essentially "dressing". It doesn't really have any impact on the end result. It doesn't matter which rider attacks first in the section based on the report... just that they attacked. Let me clarify a bit on the coding : kurtinsc is completely correct. Every part is calculated as a whole for all cyclists. So indeed, the helper is doing its job for that part. The results per rider are calculated in each part, cyclist by cyclist. So let's say that the "no tactic" rider is calculated first and the "attacker" is calculated last, you can easily see that it is not possible to "undo" his first job. In order to do this, it would require to simulate each part in a numerous of small parts, iterating over them endlessly ... BUT all of you are missing a very important detail about the role of helpers !! It's strange that, with all the extremely detailed analysis that you all have made, this very obvious thing has not been noted yet ! Another remark : in the announcement on the introduction of the "General Classification" tactic, I have finished with following statement : What happens when no GC, no Final Sprinter and no attacker is chosen ? Helpers will just relay the peloton, they will not stick with any Save Fitness rider or a "no tactic" rider.I do NOT state that a helper will NOT help a no-tactic rider, he will not "stick" with him when he drops from the peloton ...
|
|
|
Post by kurtinsc on Jan 5, 2014 21:06:07 GMT 1
Good to know Nikeboy... I misinterpreted that statement to mean not helping rather then just who the helper tries to stick with.
Do helpers still help other helpers, or is that gone?
|
|
|
Post by goof on Feb 17, 2014 20:42:03 GMT 1
Today something strange happened in the Giro. My final sprinter finished in the peloton after the attackers with one helper for the mountains. The helper finished 2nd in the group, while my sprinter finished 15 places behind him. I know that his mountain skill is not great, but I am sure that he should have finished higher than he did. He has 17 on MO+Sprint and he had 95%...
|
|
|
Post by rarau on Feb 17, 2014 22:00:35 GMT 1
Today something strange happened in the Giro. My final sprinter finished in the peloton after the attackers with one helper for the mountains. The helper finished 2nd in the group, while my sprinter finished 15 places behind him. I know that his mountain skill is not great, but I am sure that he should have finished higher than he did. He has 17 on MO+Sprint and he had 95%... helper is 7. Blijleven ? 22. Osee same time Bikkels is final sprinter? Standings after part 14 19. Kobori 4 04:53:14 20. Charbonneau 5 04:53:50 24. Blijleven 5 04:53:50 25. Romero 5 04:53:50 26. Mateescu 5 04:53:50 27. Osee 5 04:53:50 your players was not in peloton before last part. Their are in other group AFTER peloton. On last part this 2 cyclists having good mountain finish with the peloton. Cyprien Blijleven (1815) 93% Coupoussamy Osee (4595) 90% * Time Trial influence in the mountain parts. Time trial will become more important in the mountain parts. In flat parts, cyclists in a group can draft behind other riders so they can, almost freewheeling, produce the same speed as the group. In these cases it is not important that you can paddle with a constant pace (time trial). However in the mountain, this draft is barely applicable, so each rider needs to produce a certain pace (time trial). This is why TT becomes more important in the mountain parts. * Sprint influence. In the current engine, sprint is a too dominant factor in each intermediate and final sprint. For mountain and hill sprints, the main skill will have more influence.Tell us all their skills and we could give you an answer decimals too and experience. the other 2 cyclists have NO TACTIC?
|
|
|
Post by goof on Feb 18, 2014 13:44:26 GMT 1
He's got a low 7 on mountain, 10 on sprint and 3 on tt, so his mountain is rather low. The helper has got 11 on mountain and 4 on tt and sprint. So now you know. I don't know their decimals on these skills. But it's true that the influence of sprint is diminished for mountain terrain but that much? He would've done so much better previous season. And the other two cyclists were helpers too for the flat terrains.
|
|