I’d like to present my own small prediction project, the QCA prediction. Therefore I used subjective data, to mix them with formulas and grounded data to create a mixture of subjective and objective impressions. The process of this prediction has taken place independently from any other preview work.
At the end of the prediction stands a quotient, after that the teams are ranked for the final prediction table. This quotient has been created by different correlations of variables, basically classified in three categories:
Leaders vs. Depth
Focus Riders vs. Hybrids
Stats vs. Surroundings
Of course not every variable has the same influence. Examples of variables are “Stage Race leaders”, “Hilly cobbles”, “breakaway affinity” and “Calendar adaptability”. Also, the possibilities of excluding terrains as a standout-example for tactical possibilities have been taken into account; as one of the major focuses was to offer an approach that rewards depth better than most statistical methods do.
Like this the quotient should include the complete levels of MG result influences; at the end the approach is meant to include all major and most of the minor influences on the final outcome. Of course the quotient can not display all the facilities of MG rankings, but it tries to come as close to it as an approach is able to come.
The quotient itself should be pretty easy to understand. 100 is the best score, 0 is the worst score. Teams rank after it. The second number in the table is the variance. Variance should display on how realistic it is, that the outcome differs away from the suggestion the quotient offers. It reaches from 5 to 15, while 15 is the highest. Hand in hand with the variance stat the option to rank teams according to own differing thoughts higher/lower within the variance range is offered, so that especially changes between teams very close to each other are not locked out by the quotient.
On the example of Ferrero - Samruk you can see that a little closer. Because especially their leaders are very hard to predict in their score (due to their hybrid roles, shaky probable calendar races and sketchy AI treatments) they have a significantly higher variance than many other teams. Adding/Taking away 7.5 points in either of the direction would give a completely different result. Other teams like f.e. Fablok have a far lower dynamic in their possible placement outcomes with their low variance, that is explainable by their leaders that are easier to predict in terms of race expectations.
At the end of the ranking there is the rider name of one rider of the team. This is meant to be the rider causing the biggest variance, as he has the highest “boom or bust” potential of the teams’ significant scorers. These riders often decide how successful a season of a team may be, and should have a stand-out spotlight on their performances, as they can decide about promotion/relegation often by themselves.
1.
Generali - EDF
65,5
10
Roglič
2.
Orange Pro Cycling
64,5
5
Quintana
3.
Newton Foundation
63
7,5
Tsatevich
4.
Podium Ambition
62,5
7,5
Thomas
5.
Berg Cycles
61,5
7,5
Silvestre
6.
Ferrero - Samruk
61
15
Kolesnikov
7.
Grieg - Eftel
61
5
Rowe
8.
SPAR - Shimano - SCG
60,5
5
Guerao
9.
Fablok - Bank BGZ
58
5
Guardini
10.
Minions
56,5
5
Albert
11.
Team UBS
55,75
5
Schelling
12.
Kraftwerk Man Machine
54,5
5
Kreder
13.
Indosat Ooredoo - ANZ
54,5
5
Goss
14.
Eritel - Sonatrach
54,25
7,5
Debesay
15.
Meiji - JR East
53
5
Zaini
16.
Novatek-Panarmenian.net
50,25
15
Holloway
17.
Valio - DeLaval
50,25
10
Lo Cicero
18.
Carlsberg - Danske Bank
50
12,5
Juul-Jensen
19.
Isostar - Adriatic
49
15
Brändle
20.
Netia - Vónin
48,75
10
Sanchez Gil
21.
Lierse SK - Pizza Ullo PCTeam
47,5
5
van Asbroeck
22.
MOL - OMV Petrom
45,25
12,5
Di Maggio
23.
Haute Route - Mavic
39,5
7,5
Keukeleire
24.
Compal-Merida
39,5
5
Sergent
25.
Euskaltel
38,75
10
Cornu
26.
Team Ticos Air Costa Rica
34
7,5
Chamorro
The ranking surprisingly is not lead by Orange as in any other statistical prediction, but instead by Generali. They profit from being the best TTT team and basically having kept their last years squad adding the strength or a trained Roglic plus a significantly improved Edward Theuns, that with the hilly cobbles affine calendar should also score a lot better. Especially TT randomness and the race calendar of Roglic give the team a pretty high variance, which means that they could well also end up in 5th.
A safer bet on the podium may be betting favourite Orange with the big numbers on their leaders. Lacking depth on every terrain, any useful breakaway riders and potential in hybrid races make them one of the easiest to predict their final score, but also cost the final upside needed for the title. Similar situation for third predicted Newton Foundation. They have a clear leader in van Garderen, but then two upside captains in Tsatevich and Maksimov which could really make up for the needed surprise points.
Afterwards a group of 4 teams that couldn’t be more different from each other. Podium Ambition in 4th with their specified approach are clearly benefitted by their great cobbles depth and also perfectly fitting race calendar. The upside potential in any sprint does the trick. Grieg has a similar approach, but their loss of valuable cobbled and TT depth may cost them a direct promotion spot. Even more thinking about the fact that EBHs support is not the best and they completely avoid uphill riding. Ferrero instead, as already mentioned, found some perfect aggressive riders that should score in various different terrains. Thus they not only are a solid promotion candidate, but with a great season could even have chances to be in the title race. Berg Cycles are the most solid team from that range. Promotion is somehow almost guaranteed, but they should not have what it takes to compete for the title, as the big leader lacks. The incredibly deep roster on any terrain besides TT makes them so solid.
Behind those there is a long list of very solid mid-table teams. While Fablok and SPAR look very well settled for a lower Top10 and thus title hopes with their very balanced squads, other teams will be either happy (Minions, Kraftwerk) about their solid almost guaranteed mid-table save spots, or more unhappy as they miss the upside riders to aim towards promotion ambitions (UBS, Meiji, Indosat). Eritel may be the most interesting team of that range, as they can play the triumphal card of having a very deep roster that should keep them away from relegation safely even though they are missing the scoring leaders. Some climbing potential is there, mostly if they use their planning advantages well.
From Novatek on the danger zone really begins. Novatek, Valio, Carlsberg, Isostar, Netia, Lierse and MOL are all very close to each other, and on top have a high variance. That means, that all of these teams can either turn their season into a safe mid-range one, or end up on the wrong end of the relegation battle. Out of those only Lierse looks pretty solid in terms of predictional points, but their solid level is one that may be a danger one. Teams like Novatek and Isostar do not have the best of captains, but they can try to build on their riders grabbing the points in races that are not clearly favoring one type of rider. But this hybrid approach can also end up in a disaster at the end of the season, if those crazy races turn the wrong way for those.
Carlsberg has the same issue of lacking leaders, but they make use of their outstanding depth both on specialization level and on breakaway riders, as they can send 8 fitting riders to almost every race. If the depth and breakaway points stay away then it can also turn the very wrong way. Valio should be a little more save with their clear leader structure. They also kind of rely on a good breakaway squad, but if Lo Cicero halfway performs as last season they should be more solid than the others. Lierse is pretty similar, though on a lower level. The roster has slightly improved, but last year van Asbroeck had an incredible season. He needs to repeat that if they want to stay safe.
Netia is a team full of question marks. The leaders should all score solid, but behind that Sanchez needs to find another vintage year to keep them safe. If not, Kwiatek needs to score big in the hybrid races - because the main terrains of the team, hills and sprints, have improved the most in PCT this year. They are definitely missing both the depth and the breakaway squad to cover leader failures. MOL sits on the first relegation place, but indeed could leave that at some point. None of the leaders has the quality to win a race, but at least they have plenty of them. A di Maggio finding his mojo could save their season. Otherwise they have to rely on the attacking riders deeper in their roster and need the surprise effect on their side.
4 teams are really in danger according to this quotient: Haute Route, Compal, Euskaltel & Ticos. They are significantly behind the other teams. Compal has some leaders, but they are all no big deals. What they are really lacking is riders that don’t just hang on, but go for themselves, or a sprinter to be fed by the numbers of decent leadouts. Almost sure a lucky punch, most likely from Sergent in a big ITT, is needed to have a good position against relegation. Haute Route looks better in terms of their winning potential, but they have kind of the same issue. A stage racer as a leader that sadly is not even a Top 10 stage racer. They have to rely on one of their sprinters going big on differently raced sprinters classics.
Euskaltel has the support. They arguably had the biggest steal in the transfer period. But they still are in big danger; the problem is that they are mediocre in every terrain, but don’t have one really big scorer. Thus a fitting schedule is hard to pick. Cornu in theory should be the factor being one of the best time triallers in the division, but when it mattered he constantly proved to be different. Ticos however are sad to watch. They built an outstanding deep attacking roster. Give them one more season and they have a big-time PCT time. Sadly they won’t have the season. Their great squad lacks any sort of leader. There is just no rider that is close to score something when there is no mountain in the race.
I very much hope you liked the preview. Feedback and comments are very much appreciated.
Nice preview Croatia! Very interesting in that you have the first prediction without Orange in front. As for your prediction for us, it is pretty much in line (albeit on the lower end) of what other predictions had. Interesting analysis all around though, thanks.
Bushwackers wrote:
Nice preview Croatia! Very interesting in that you have the first prediction without Orange in front. As for your prediction for us, it is pretty much in line (albeit on the lower end) of what other predictions had. Interesting analysis all around though, thanks.
Thank you very much! Your season will be decided from what comes behind Amador. Because Andrei is the best even clearer this year. I'm honestly struggeling to find races where the likes of Schelling, Reinhardt and Aregger score big, and this calendar discrepancy surely have costed you (some) places beneath lacking a clear second leader.
I can see you well finishing a little bit higher though, but in my eyes you may lack a 78-79 TT rider for that to make your TTT squad the strongest.
It's true Van Asbroeck should be a key rider here, but I think the sprint field went up by a lot. His backup stats are amongst the best in the PCT, but he lacks a bit of speed. Personaly I think it will be Waeytens who will make things more interesting but we'll see I guess.
Very nice analysis Croatia! Using parameters beyond just stats would've taken quite a bit of computations and adjustments, so impressive work. Not a part of PCT so can't comment on the nitty-gritties too much. Would love to see something similar for CT
I think you're fairly spot on regarding us here. Last year our sprint train performed a lot better than anyone expected and we're betting on that to happen again this season.
If Lo Cicero doesn't perform like last year we're in for a tough one...
Also interesting remark regarding the breakaway affinity as that was a clear-cut strategic decision made when shipping off Anacona and not going after a new mountain leader. Time will tell whether or not it pays dividends
I would ask all the people from the Balkan to contact me so that we can create our community.
List of Balkan countries:
- Albania
- Bosnia & Herzegovina
- Bulgaria
- Croatia
- Greece
- Kosovo
- Macedonia
- Montenegro
- Romania
- Serbia
- Slovenia
- Turkey
@Ollfahrd: Well, but luck on the cobbled stages can easily make up for 200+ points. Waeytens can't really score better with luck in my eyes, it's more of a determined than a range scorer. And yes, I seem to rate your team a lot higher than others. But well, it grown better and you trained Waeytens (though I surely would've tried to give van Asbroeck that 81 sprint)
@Abhishek: Thank you very much! Indeed it took countless of approaches to figure out a fitting formula, and it values all parameters individually different and even in their relations to each other, which I'm kind of proud of.
@Tamijo: Thanks! The variance was a late idea of mine, but I thought that it raises the understandings of the preview and it's relations including the assumptions you can take out of it. Quite happy with the result, not only because it offers excuses for failed predictions on some teams
@TMM: Yes 4th, and even more valuable: Even with the variance alignment quite comfortably into promotion area. To be fair though, you also had the easiest transition to PCT with 3.5 worthy scorers Still I rate your spot on offseason work on strengthening your core intelligently very high.
@Atlantius: Some nice words and good to hear that you agree with me on your team. I very much recognized this even more attacking flaw of yours in the offseason, and I really like the decision. Also: Seeing your team in attacks just feels so right anyway.
@Djordje: Well, in a sparer edition I can offer you https://pcmdaily.com/forum/viewthread....st_1264383. The extend of variables and parameters are not that in depth, but still the coefficient used kind of a light version of this as a basis.
@Matt: deal! can we make it "if you finish on the podium" though?
Croatia14 wrote: @TMM: Yes 4th, and even more valuable: Even with the variance alignment quite comfortably into promotion area. To be fair though, you also had the easiest transition to PCT with 3.5 worthy scorers Still I rate your spot on offseason work on strengthening your core intelligently very high.
I can only hope those depth points can work enough to keep me out of the cellar. We'll see I guess! Juul-Jensen is indeed he biggest wildcard. Is he the next De Maar? I really hope so.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
baseballlover312 wrote:
I can only hope those depth points can work enough to keep me out of the cellar. We'll see I guess! Juul-Jensen is indeed he biggest wildcard. Is he the next De Maar? I really hope so.
Whether he will be a de Maar type of rider or just a 78 cobbler with great acceleration depends more than almost any other rider on your planning of him. For you personally I hope you did a great job, as your relegation opponent I'd like to not have you outplanning me as you did in CT last year.
baseballlover312 wrote:
I can only hope those depth points can work enough to keep me out of the cellar. We'll see I guess! Juul-Jensen is indeed he biggest wildcard. Is he the next De Maar? I really hope so.
Whether he will be a de Maar type of rider or just a 78 cobbler with great acceleration depends more than almost any other rider on your planning of him. For you personally I hope you did a great job, as your relegation opponent I'd like to not have you outplanning me as you did in CT last year.
If I outplanned you, it was 100% luck. I am much less confident about it this year even. So we'll see.
RIP Exxon Duke, David Veilleux, Double Feature, and Monster Energy
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
