|
Pre-Renewals DB 2018 (v2)
|
| SotD |
Posted on 19-06-2018 13:00
|
World Champion
Posts: 12188
Joined: 29-11-2006
PCM$: 2980.00
|
You use him wrong 
|
| |
|
|
| quadsas |
Posted on 19-06-2018 13:15
|
Small Tour Specialist
Posts: 2518
Joined: 18-01-2013
PCM$: 300.00
|
9 Lithuanians in the db make me a sad boy |
| |
|
|
| Aquarius97 |
Posted on 19-06-2018 13:21
|
Grand Tour Specialist
Posts: 4191
Joined: 13-09-2015
PCM$: 300.00
|
It's 9 Lithuanians who currently have a contract with teams. FA aren't added to this DB
|
| |
|
|
| Roman |
Posted on 19-06-2018 14:16
|
Grand Tour Specialist
Posts: 4386
Joined: 29-05-2007
PCM$: 200.00
|
In my opinion the new OVL system needs changes.
Sprinters are overrated. My rider Habtom went from 74,21 to 75,69. He has a higher OVL than riders like Vlatos or Arndt. Habtom was 406th in PT rankings. These two were 87th and 75th. In 2016 they were 83rd and 51st, while Habtom 281st unmaxed. Or Daniel Vesely: he was 183rd best rider in the DB last year, now he is 97th. Has higher OVL than riders like Keinath, Olivier or Borges. Vesely was 112th in PT rankings and I believe it was quite an overperformance. These three were 33rd, 51st and 52th.
Also, cobble riders and those with a higher sprint especially are also overrated. Totally ignoring insane Bewley's OVL, what about GVA? Went from 80,12 to 79,93 after declining! Kristoff went from 77,89 to 79,59!? Blythe from 78,48 to 79,22. Trentin from 78,53 to 79,27. Saber from 77,02 to 78,66. If you take a look on the results from last year while ignoring GVA, they were in PT rankings 48th, 59th, 35th, 82nd. On the other way 22nd Ginnani has lower OVL than all of these. Tim Wellens, who was 38th has more than 1 point lower OVL than even the worst of these riders - Saber. Even 7th Alarcon has just a little bit higher OVL than Saber. The year before it was 38th, 50th, unmaxed 185th vs N/A, 24th, 20th.
In my opinion big majority of MG cobble races are hard enough that sprint attribute almost does not matter in them in the end, mainly because 99% of riders don't have any power left for a sprint. That 1 percentage is Bewley and GVA and even with them you can say that they would still be the biggest favourites even with having SP stat around 70. Just take a look on results of sprints in cobble races. Sprint attribute almost does not matter, more important IMO is acceleration. HIL and DH also makes quite a difference. In this year's Roubaix we have seen both Trentin and Blythe being outsprint by 65 sprinter Sulimov. Or in Ronde Senechal and Vanspeybrouck beating Trentin. And also if a cobble rider has a good sprint - it already makes a difference as it is likely his 2nd best race category.
Or well, tell me: should Zhupa (ranked 215th in PT last year, 299th unmaxed) have a similar OVL as Prevar (131st, N/A), Kittel (79th, 183th) or Furdi (129th, 189th)? Should Gough (361st, 347th in 2016) be around Malori (115th, 129th) or Aru (127th, 145th)? Boroš (194th, 194th) higher than Tratnik (106th, 233th unmaxed) Vosekalns (100th, 275th unmaxed) or Mager (138th, 208th unmaxed)?
Or in other teams: how can be Dakteris (204th, 201st) better than Koch (85th, 165th)? How is Tom David (93rd, 90th) so much better rider than Vlatos (87th, 83rd)? How is Blythe (59th, 50th) better than Kelderman (47th, 22nd)? How is Weber (349th, 203th) better than Arndt (75th, 51st)? Or what about Eric Young (394th, 155th) - is he really so much better than Cataford (162nd, 170th)?
And with sprinters it is all the same. Emerson Santos a higher OVL than Vlatos (87th, 83th)? That Emerson Santos who has almost the same attributes as Te Brake (449th, 480th), Aberasturi (350th, N/A) or Keough (347th, 515th)? Wippert (421st, 218th) better than König (96th, 121st)? Feiereisen (354th, 466th) vs Kittel (79th, 183th)?
Maybe top 5 sprinters are ranked quite ok, but that should not result into other sprinters being overrated. And similarly one Bewley should not mean all cobble riders get too high OVL. I don't really believe Bewley is that good. Spilak was 10 points behind him and Bewley was behind Schleck in rankings in 2016, with Phinney being around 200 points behind him in both years. And well, is Bewley really 1.4 points better than Phinney when both of them have 83 as their top 2 main attributes and both of them have really good support stats? I can see the case for Bewley being the best rider (or a joint best rider with Taaramae), but he surely is not that good so the gap between him and Phinney is the same as a gap between Phinney and a rider like Lecuisinier.
|
| |
|
|
| DubbelDekker |
Posted on 19-06-2018 14:28
|
Small Tour Specialist
Posts: 2633
Joined: 20-04-2008
PCM$: 200.00
|
Roman wrote:
If you take a look on the results from last year while ignoring GVA, they were in PT rankings 48th, 59th, 35th, 82nd. On the other way 22nd Ginnani has lower OVL than all of these.
Ginanni is 33 now, he declined.
|
| |
|
|
| Roman |
Posted on 19-06-2018 14:46
|
Grand Tour Specialist
Posts: 4386
Joined: 29-05-2007
PCM$: 200.00
|
DubbelDekker wrote:
Roman wrote:
If you take a look on the results from last year while ignoring GVA, they were in PT rankings 48th, 59th, 35th, 82nd. On the other way 22nd Ginnani has lower OVL than all of these.
Ginanni is 33 now, he declined.
You are right, my mistake. But it is the same case with for example Lachlan Morton. Way lower OVL than all of three mentioned cobble riders, while being higher in PT rankings than all three last year. Gastauer or Adam Yates as well. Adam Blythe is nowhere near being a better rider than Jose Alarcon.
|
| |
|
|
| knockout |
Posted on 19-06-2018 15:26
|
Grand Tour Champion
Posts: 7735
Joined: 21-12-2010
PCM$: 400.00
|
Still have to take alonger look at this but i have a question:
What was the intended reaction to this? Laughing? Choking? Pulling an Alak? Long rants? 
A Big Thank You To All MG Reporters! |
| |
|
|
| Heine |
Posted on 19-06-2018 15:28
|
Grand Tour Specialist
Posts: 4116
Joined: 08-04-2007
PCM$: 200.00
|
Havent been able to look at the db, but isnt ovl pretty much updated every season?
|
| |
|
|
| SportingNonsense |
Posted on 19-06-2018 15:32
|
Team Manager
Posts: 33046
Joined: 08-03-2007
PCM$: 200.00
|
Certainly sprinters are ranked higher now, but if the consensus is that the weaker sprinters are universally too strong, then this can still be changed. Indeed Roman has some convincing examples, in riders just in my own team. Constructive feedback is always welcomed!
If analysing old vs new though, please do focus on relative OVL rank. The change in OVL value is somewhat irrelevant, in my opinion, as the changes affect the scaling in many ways. Attempts are made to put the values on a comparable level, but in this instance it is more true for the top riders, than the rest.
Anything that uses the OVL will be adjusted accordingly, such as RD and Renewals impact, as they always have been when OVL is changed.
|
| |
|
|
| the_hoyle |
Posted on 19-06-2018 15:58
|
Grand Tour Champion
Posts: 7651
Joined: 28-05-2009
PCM$: 200.00
|
Great work compiling this all together! I can look at working on the NC winners list when i can have a 10 mins or so each night
.: Manager of :.
.: My Awards :.
 |
| |
|
|
| Ad Bot |
Posted on 22-11-2024 18:53
|
Bot Agent
Posts: Countless
Joined: 23.11.09
|
|
| IP: None |
|
|
| baseballlover312 |
Posted on 19-06-2018 16:01
|
Tour de France Champion
Posts: 16429
Joined: 27-07-2011
PCM$: 10438.70
|
I'm just gonna observe the discussion here for the most part because I trust that those in charge (and most other managers) have a better knowledge and understanding of the formulas than I do.
Seems like basically all of my overalls have gone up, which is fine. Juul-Jensen and Appo went up a bit, which makes sense. Others like Nolf going up at all is a little shocking to the formula when his rider type proved so unsuccessful last year. And I definitely see that lower sprinters getting very large overall hikes. On my team, AKA now has a higher overall than JJ at 80/77, and Bertillsson and Ranneries also got pretty big hikes to look like leaders. But if other formulas have been adjusted to compensate, I don't think it's much of a problem. We just need to iron out the kinks.
An important thing to note is that OVL values change all the time! I went and looked back at my first 2014 DB from when I joined, and clearly the formulas were massively different. Fighters and domestiques had way higher overalls back then and that was changed without many even noticing or caring. I think this will all work out fine.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy |
| |
|
|
| roturn |
Posted on 19-06-2018 16:17
|
Team Manager
Posts: 22246
Joined: 24-11-2007
PCM$: 3900.00
|
the_hoyle wrote:
Great work compiling this all together! I can look at working on the NC winners list when i can have a 10 mins or so each night
No need any longer. SotD already did this earlier today. |
| |
|
|
| SportingNonsense |
Posted on 19-06-2018 16:24
|
Team Manager
Posts: 33046
Joined: 08-03-2007
PCM$: 200.00
|
knockout wrote:
Still have to take alonger look at this but i have a question:
What was the intended reaction to this? Laughing? Choking? Pulling an Alak? Long rants?
No such thing as an intended reaction.
Ideal reaction. Everyone likes it, we move on
Expected reaction is that if anyone doesnt like the changes, they'll say why. Probably followed by a discussion, and if it's a valid point, then further changes can be made.
More extreme reactions are certainly unnecessary, and are hardly helpful.
I definitely think this OVL is better now in general, but it will not be better for all riders - and if it has gone too far for some cases then I'd expect this thread to identify that.
|
| |
|
|
| alexkr00 |
Posted on 19-06-2018 16:56
|
World Champion
Posts: 13915
Joined: 05-08-2008
PCM$: 300.00
|
Sprinters OVL are definitely too high.
Looking at my team, Grosu is currently the best 35th rider under contract which doesn't exactly feel right to me considering how he was 70th in the PT rankings last year. Yes, with better planning and a bit of more luck he could score better than last year but I still feel like that OVL is too high for him.
While I can still see a point for Grosu's OVL being that high, Te Brake's is just ridiculous. He is my 6th best rider based on OVL now and looking at my team that doesn't feel right. How is his OVL higher than Brambilla's? I don't see any way this can be explained at the moment.
And speaking of Brambilla (even weirder cases have been pointed out) how is his OVL lower thank Sinkeldam given their stats.
|
| |
|
|
| Roman |
Posted on 19-06-2018 17:11
|
Grand Tour Specialist
Posts: 4386
Joined: 29-05-2007
PCM$: 200.00
|
One additional thing to put here: I think the balance is quite fine between climbers and puncheurs, it is a really nice thing to see that ACC matters way more now. There is also a clear improvement to better reflect a quality of the strongest riders on flat.
But on COB and SP - maybe it can work at least partially from a point of view that a higher OVL = less available RDs for them, so worse sprinters can possibly have more chances to score some points and so to justify their higher wages. But that would also mean less scored points for the best sprinters than in 2017, so even those top riders end up being overrated in comparison to their PT rank.
|
| |
|
|
| Yellow Jersey |
Posted on 19-06-2018 18:06
|
Classics Specialist
Posts: 2809
Joined: 16-04-2013
PCM$: 900.00
|
I may be wrong but I believe there's something wrong in the XP in two of my riders, Bonaventure Uwizeyimana and Daniel Rinner were both level 4, with 22 and 0 XP respectively, and those numbers have remained the same. My mistake or needs a little fix?
|
| |
|
|
| Aquarius97 |
Posted on 19-06-2018 18:11
|
Grand Tour Specialist
Posts: 4191
Joined: 13-09-2015
PCM$: 300.00
|
Well, in CT is impossible to get Lvl4 XP points unless you ride in HC race, which you didn't last season
|
| |
|
|
| knockout |
Posted on 19-06-2018 18:16
|
Grand Tour Champion
Posts: 7735
Joined: 21-12-2010
PCM$: 400.00
|
I did not look long into the DB so this is far from a complete analysis of the new ratings and rather quickly used excel to throw out some diagrams but hopefully it raises more discussion on the OVR.
For the purpose of simplicity I'm only looking at riders who were in PT last season and I exclude everyone over the age of 32 to filter out declines and everyone unmaxed. Newly maxed riders will impact the data a bit but the impact shouldnt be too huge because they impact each category of riders.
2017 PT Points by Type of rider
I used a very simple method of picking the type of rider: The highest stat of Mo, Hi, TT, CB, SP decided the specialization. I know it's not as simple as that but its sufficient for what i want to show for now:
x axis: new OVR
y axis: PT points 2017
I admit that's a bit too crowded to see a lot so let's zoom in a bit. I cut off everything under 74 OVR because those riders should generally be in the 50k wage area so a bad formula for those doesnt matter much and we get more "didnt use full RDs" effects and stuff like that. I also cut out everything above 78 OVR because then it's gets too small a sample size to really be interesting (although i might want to point out that even there it is obvious that most positive outliers are climbers).
Let's zoom in:
We can see a trend appearing but let's add some visual aid: For each category I add a rolling average (5 periods) to get an easier comparison of the different specialisations:
Climbers have by far the most points per OVR by the new OVR calculation. Hence their OVR must be highered (or others be lowered) and I'm obviously not only saying that because of having little climbers.
Sprinters score not even half as many points as riders of other specializations with a similar OVR. Mark Dzamastagic is the only sprinter to be above the rolling average of climbers and he needed absolute fluke results to get there. Meanwhile there is just one climber and hilly rider (no TT / cobbler) below the moving average for climbers.
tl;dr The sprinter OVR is gamebreaking at this point. Climber OVR has to rise in comparison to other specialisations.
I hope to find some time this week to look further into other aspects but this cannot be the final formula. That said it would be interesting to find out the exact formulas
EDIT: When i predicted each rider's points for the season in my season previews one of the biggest adjustments i had to make was to reduce the points for (below top tier) sprinters a lot and raising the average of sprinters that much is ridiculous.
Edited by knockout on 19-06-2018 18:29
A Big Thank You To All MG Reporters! |
| |
|
|
| Ripley |
Posted on 19-06-2018 18:29
|
Classics Specialist
Posts: 3341
Joined: 25-11-2014
PCM$: 300.00
|
Nice work, knockout!
|
| |
|
|
| Yellow Jersey |
Posted on 19-06-2018 18:33
|
Classics Specialist
Posts: 2809
Joined: 16-04-2013
PCM$: 900.00
|
Aquarius97 wrote:
Well, in CT is impossible to get Lvl4 XP points unless you ride in HC race, which you didn't last season
Ah yes that's what I was missing  thanks!
|
| |
|
