Like last year's thread by Jandal I will start with a figures based exercise for PCT. Like last year I refuse to call it a prediction:
Top 3
Top 5
Top 10
TTT
Score
TT factor
1
Xero Racing
78,81
77,67
75,83
72
77,44
84,64
2
Fastned
78,3
77,57
75,96
70
77,28
84,28
0,16
3
Los Pollos Hermanos
78,13
77,58
75,95
75
77,22
84,72
0,06
4
Assa Abloy
78,14
77,29
76,1
72
77,18
84,38
0,04
5
DuckDuckGo-Everesting
78,07
77,38
76,06
72
77,17
84,37
0,01
6
Lierse
78,08
77,58
75,81
74
77,16
84,56
0,01
7
Tryg Gobik
78,29
77,1
75,87
73
77,09
84,39
0,07
8
Sony Force India
77,88
77,12
76,01
73
77,00
84,30
0,08
9
Popo4Ever
77,7
76,8
75,69
72
76,73
83,93
0,27
10
Lampre
77,24
76,84
76,07
69
76,72
83,62
0,01
11
Jura
77,16
76,84
76,14
77
76,71
84,41
0,00
12
Colombini Cycling
77,44
76,72
75,77
71
76,64
83,74
0,07
13
Indosat
77,56
76,7
75,62
75
76,63
84,13
0,02
14
Bralirwa-Stevens
77,77
76,69
75,4
75
76,62
84,12
0,01
15
Team UBS
77,14
76,56
75,73
75
76,48
83,98
0,14
16
DK Zalgiris
77,15
76,54
75,7
72
76,46
83,66
0,01
17
Zwift Pro
76,66
76,41
75,81
72
76,29
83,49
0,17
18
McCormick
76,86
76,28
75,47
75
76,20
83,70
0,09
19
Kraftwerk
76,81
76,24
75,54
73
76,20
83,50
0,01
20
Sauber
76,43
76,19
75,47
73
76,03
83,33
0,17
21
Trans Looney Tunes
76,4
76,12
75,47
73
76,00
83,30
0,03
22
Spark BNZ
76,57
75,97
75,11
74
75,88
83,28
0,11
23
Genii Hyundai
76,13
75,67
75,01
74
75,60
83,00
0,28
24
Glanbia
76,33
75,38
74,22
76
75,31
82,91
0,29
Basically the score is the average of the top 3, top5 and top 10 of each team. This was broadly accurate last year for the CT except for my own team whose TT power wasn't taken into account.
So what happens if we sort by a powerful TT factor just for fun:
24/02/21 - kandesbunzler said “I don't drink famous people."
15/08/22 - SotD said "Your [jandal's] humour is overrated"
11/06/24 - knockout said "Winning is fine I guess. Truth be told this felt completely unimportant." [ICL] Santos-Euskadi | [PT] Xero Racing
roturn wrote:
Thank god you can avoid a lot of the TT stuff in PCT.
I hope 22 other teams are thinking alike, I'd love to monopolize those races
But I'm afraid that despite the likes of Eastman, McNulty, Reis or Powless going to ToA, especially Dunbar will still be there to cause me lots of nightmares - and obviously the Pollos and Assa Abloy duos...
I'll definitely take that second non-prediction for my team, though - although 4th place doesn't really look realistic. But even with the first variant we should be safe, which clearly is the primary goal for this year.
Thanks for starting the thread - I'll post some templates again once the micro jerseys are there (no pressure though!). If we get enough predictions, I'll happily put together community predictions for PCT and PT, as Ezee seems to have "stolen" that one for CT
Another lazy prediction, this time for the PT. I’ve concentrated entirely on numbers here for two reasons:
1) I’ve no clue whatsoever regarding the PT, so there’s no point in trying to make a prediction regarding planning, race favourites or past seasons.
2) So it’s totally objective, because it’s mathematic (and I have to know, because I’m teaching Maths ).
How does it work?
The idea behind it is to value the top riders as well as depth in each terrain.
To achieve this I’ve simply sorted all PT riders in regards of the five main stats MO, HI, TT, CO and SP. Then I’ve assigned 50 points to the best rider in each stat, 49 to the second best, 48 to the third best up to 1 point for the 50th best and added those points for each team. In case of stat ties I’ve sorted them by the sum of the three energy stats, then a second important stat for the respective kind of rider (TT for MO; MO for HI; PR for TT; FL for CO and AC for SP) and as a final tie breaker I used the OVL. So not much thinking, but some adding – a not quite so lazy prediction.
My PCT "Prediction". As is tradition a bunch of qualifications. This is based purely on a rider level prediction of scoring, so no consideration of team make up. If you have 9 sprinters the algorithm assumes they all get enough RDs. Also, doubt the algorithm is that precise so the margin of error is pretty big.
So for example there might be some reliability about the top 4, given the gap to 5th but would put little credibility in the sequence from 1 to 4 since they are all within 123 points. And then there is all the other unpredictability introduced by the race planning aspects of PCT
This is to a significant degree an OVL based prediction so not surprisingly a lot of crossover with Mac's first exercise.
In the spirit of being lazy, I've created a prediction method that only requires you to think about two teams at the same time. For every division, you simply get 20 randomly chosen team matchups and decide which of them will have the better season (i.e. finish higher in the rankings). The more people participate, the more matchups will be decided and the more representative the ultimate result will be, as teams will be ranked by the percentage of matchups they win. You can do each one multiple times to get different matchups, too, if you're bored
Used metrics similar to Conglomerate preview mechanisms to craft a CT prediction. This one is only based on rider quality, I don't know the calendar or planning mechanisms in detail.
@Croatia Quite surprised to see Caja this low. Some glaring weaknesses for sure but compared to teams like SEE Turtles or Adidas (no offense to those in particular, could've named a couple others, too), I'd expect them much higher. It's an interesting contrast to Ekoi in second place, who have similar holes (barely any scoring potential in sprints, time trials and cobbles) and surely have more depth, but aren't that much stronger with their top riders. Lots on the shoulders of Paret-Peintre and Teuns to get them to second place. I would be skeptical if I didn't put as much stock into your MG savvyness as I do
cunego59 wrote:
@Croatia Quite surprised to see Caja this low. Some glaring weaknesses for sure but compared to teams like SEE Turtles or Adidas (no offense to those in particular, could've named a couple others, too), I'd expect them much higher. It's an interesting contrast to Ekoi in second place, who have similar holes (barely any scoring potential in sprints, time trials and cobbles) and surely have more depth, but aren't that much stronger with their top riders. Lots on the shoulders of Paret-Peintre and Teuns to get them to second place. I would be skeptical if I didn't put as much stock into your MG savvyness as I do
I understand, this is mostly a gut feeling and minor metrics list, but I indeed rate the captains and composition of Ekoi very highly. I don't really like the stat distribution of Caja's captains, that's why they're low. But of course a lot of taste. To be honest CT is very strong this year with a lot of very well built new teams which I really like. Quite impressed on what they did with a low level Free Agent market and limited options.
Like I did in the previous years, I'm planning to do community predictions once again - at least for PT and PCT, as CT might be covered by Ezee already.
However, if you want to have it even simpler, I've prepared the templates for you for each division - just quote this post, remove everything you don't need, and fill in your rank predictions for each team. You can do one division, two, or all three of them. And you can add some columns if you want to add some explanatory data (or comments) - feel free, it's there to be used
In the end, I'll collect all predictions I can find in the available preview threads and try to come up with a "community" prediction for each division. The more participants, the more fun it will be
And big thanks once more to alex for the great micro jerseys!
A pattern emerging for McCormick that says we will be somewhere between 17th and 18th. Which I would be very happy with. Survive this year and build from there.
So, can we just take kb's PCT prediction and not even do the season? Would save a ton of work for reporters
Whereas I like the fact that this model rewards depth to a certain degree - with the Top 50 of each category getting some points - I guess the biggest flaw is that it doesn't reward the top riders enough. For example, looking at the PT prediction, I doubt Puma will finish so far down. Last year, Herklotz and Phinney were just 500 points above the rest - and with the top climber getting 50 points and #5 still getting 46, I think that difference doesn't reflect the actual scoring potential. So I don't think Puma, Evonik or Moser will finish that far down.
It's still an interesting prediciton, though - for example, I knew Tinkoff were really strong in TTs and Cobbles, but I totally forgot they also have a strong (although pretty one-dimensional) puncheur department Something else I see is that although ELCO - ABEA aren't top in any domain (except for Coquard likely rocking the sprints again), they are solid everywhere and don't have a truly weak domain, being the only team having 50+ points in every category. Thanks for doing these calculations - just like kb, I just appreciate numbers (and interpreting them )
In terms of PCT predictions, I'm pretty surprised to end up in the lower Top 10 in most of them - or just Top 10 in all but Mac's first prediction. I was rather expecting to battle against relegation or - if we have a good year - fight for 15th, but a Top 10 finish would definitely be great
What I obviously like most is that TT number in kb's prediction - given that TT-heavy races are the easiest to get some great depth results (remember Glanbia's dominant Olympia's Tour last year), this is where we might really make our TT-heavy roster worthwile. Although we don't have any top riders except for Theuns, I hope we can get some good lower Top 10 results in many races! And on the cobbles, I'm obviously not sad at all about those five zeros (nor about three zeros in TTs).
What's also noteworthy is the maximum value of 86 in sprints - which is the only category across all divisions where there's not a single value above 100. Which probably just means the competition will be insane - and daily form could once again play a big role.
In CT, we have the only team across all division scoring 100+ in all categories, which I definitely didn't expect when just taking a global look over the lineups. The relative lack of competition in both TTs and cobbles surely helps, but it's true that Tafjord have a rather stacked roster - it remains to be seen if those riders are performing according to their main stats, though. I'd not be surprised to see Tafjord an Podium Ambition among the promoting teams, though.
Thanks again for doing this, will definitely be interesting to compare those numbers to the actual outcomes
The CT predictions indicate some clear favourites with the two Nordic teams and mostly unanimously see most of the new teams at the bottom of the ranking, but in between it seems like anything is possible - with a certain French teams ranking ranging from 2nd (thanks Croatia ) to 17th (thanks to myself ...), with some people either agreeing that we're a) clearly in contention for promotion or b) a solid midfielder or c) more of a backmarker. And I see quite some teams with similar results. I take it as a sign that the CT will see a really close and exciting season, emerging as the place to be if you like racing at its finest. Who needs PT or PCT?
@UU: I'm happy my PCT prediction proves to be totally accurate by placing you 17th, only 2 points ahead of 18th
otherwise it will be hard to recovre for next season.
