I feel like I don't have enough knowledge of how all the stats really work together, and of ManGame history, so I don't feel like writing my own preview.
Instead, I'd like to invite you all to participate in a kind of prediction game. Basically, you simply fill in for each team the final position you expect them to take. Below, you can find a template for each of the divisions. You have a table with rank, team name and comments, where the latter is of course optional.
You can try to predict all divisions, or just one or two. The more participants, the more fun it will be analyzing the results
Of course, I'll then post the (anonymous) results, while trying to formulate some words about the top and bottom candidates of each division. I won't set a deadline for participating now, as I don't know yet when the season will kick off. I wanted to try something new/different, so if it works, great, if it doesn't, I won't do it again next year
Please don't modify the order of the teams in the templates, just fill them in and send me a PM!
Just bumping this one. I've received one prediction for CT and PCT each, and even if I count the "official" previews in, it wouldn't be as representative as I'd like it to be.
So if you've got some spare time, just fill in your predictions for one or several divisions
I'll put up the preview by the end of the week - although we're already going to be in the middle of TdU by then, I guess it's still early enough. Want to finish some reports first and add my own predictions to it, so feel free to submit your ranking predictions until Friday.
Other than the predictions by PM, I'll take into account all published previews, plus some posted only on Skype (with their author's permission). If you feel like your preview shouldn't be used for this (too random, or other reasons), tell me and I'll exclude it
The season has kicked off, the first races are done - so it's now or never to post some more previews!
As announced before, the "Community Preview" is a collection of all the predictions posted on the forum (unfortunately, Leif's didn't get to the end yet - would surely have been very interesting, too!), predictions submitted by PM and some submitted only on Skype.
Overall rankings are established in two different ways:
- By average rank prediction, meaning "outliers" have some influence - the fewer contributions there are, the heavier they will weigh!
- By median rank prediction, i.e. the "middle" value of all predictions is taken into account. As there is an even number of predictions for PCT and CT, there are two "middle values", hence the average of both is taken.
- If several teams have the same average or median value, the other value is taken as a tie-breaker.
Without further ado, let's start with the 2020 PT Community Prediction:
Number of predictions taken into account: 7
Prediction by average rank
Rank
Team
Average prediction
1
Evonik - ELKO
2.14
2
Moser - Sygic
4.86
3
Isostar - Specialized
4.86
4
eBuddy
6.29
5
Gazelle
6.86
6
Festina - OAKA
6.86
7
Team Puma - SAP
7.00
8
Generali
7.43
9
Bennelong - Mitchelton
10.14
10
Aker - MOT
12.00
11
EA Vesuvio
12.00
12
Air France - KLM
12.71
13
Team Tinkoff - La Datcha
13.14
14
Volvo acc. by Spotify
13.29
15
Team UBS
13.86
16
King Power
14.43
17
Rakuten Pro Cycling
14.57
18
Grieg-Maersk
15.14
19
Aegon - Peroni
17.29
20
ISA - Hexacta
17.86
21
Indosat Ooredoo
19.43
22
cycleYorkshire
20.14
Prediction by median rank
Rank
Team
Median prediction
1
Evonik - ELKO
2
2
Moser - Sygic
3
3
Isostar - Specialized
4
4
eBuddy
5
5
Gazelle
5
6
Generali
5
7
Festina - OAKA
6
8
Team Puma - SAP
7
9
Bennelong - Mitchelton
10
10
Aker - MOT
12
11
EA Vesuvio
13
12
Air France - KLM
14
13
Team Tinkoff - La Datcha
14
14
Team UBS
14
15
Volvo acc. by Spotify
15
16
Grieg-Maersk
15
17
King Power
16
18
Rakuten Pro Cycling
17
19
Aegon - Peroni
18
20
ISA - Hexacta
19
21
cycleYorkshire
20
22
Indosat Ooredoo
21
Both methods see the exact same Top 5 teams - and even in exactly the same order. However, while the range of predictions is very small for Evonik - ELKO (1st to 4th) and Isostar - Specialized (2nd to 7th),
it is above 10 for the others - meaning that most predictions saw them really high up, with just one or two outliers.
The Top 8 teams are the same in both tables, too, as the majority seems to see a certain gap to Bennelong - Mitchelton in 9th and then again to Aker - MOT in 10th place.
At the bottom, cycleYorkshire really seem to be doomed, as all predictions had them on the last 6 spots, whereas Indosat Ooredoo - even though having the worst median value - could reach safety according to some. But again, the bottom 4 teams are the same with both calculation methods, with just the red lantern swapped.
Overall, the team with the biggest prediction range are Festina - OAKA, where the numbers vary from 1st to 19th, whereas the one team where we have the highest agreement on their final ranking region are Evonik - ELKO, with predictions from 1st to 4th.
PCT and CT predictions will come later today - stay tuned and discuss
It might happen but hopefully my leaders can carry the team enough plus lots of freedom in many races for Velits/Valls to avoid relegation.
In especially as KoM jerseys can't really be predicted but hopefully being attacked by those two.
Let's move on to PCT, where we have by far the most predictions available - will this also mean that the community prediction for this division is the most accurate? Well, experience shows that it's probably the most difficult to predict as well, so we'll see.
Let's jump straigt into the numbers for the 2020 PCT Community Prediction:
Number of predictions taken into account: 14
Prediction by average rank
Rank
Team
Average prediction
1
Mapei
2.14
2
Farfetch Pro Cycling
4.50
3
Desigual
5.71
4
Repsol - Netflix
6.57
5
MOL
6.57
6
Team Popo4Ever p/b Nemiroff
7.07
7
Polar
8.50
8
Lierse SK - Pizza Ullo PCTeam
9.86
9
Berg Cycles
10.50
10
Duolingo
10.93
11
Xero Racing
11.50
12
Podium Ambition
13.29
13
Bakkafrost
13.64
14
DK - SVA Zalgiris
14.14
15
Minions
14.29
16
Fablok - Chocolate Jacques
15.07
17
Swisslion Cycling Team
15.14
18
Kraftwerk Man Machine
15.71
19
Valio - Viking Genetics
18.14
20
Adastra N Hell Energy Cycling
18.36
21
GCN Racing
19.29
22
Aramco DP
19.43
23
Nordstrom - CA Technologies
19.93
24
Campari Asahi Procycling
23.21
25
Azteca - NBCSN
23.36
26
Andorra Cycling Project
24.14
Prediction by median rank
Rank
Team
Median prediction
1
Mapei
1
2
Farfetch Pro Cycling
3
3
Repsol - Netflix
4
4
Team Popo4Ever p/b Nemiroff
5
5
Desigual
5.5
6
MOL
6.5
7
Polar
8
8
Lierse SK - Pizza Ullo PCTeam
9
9
Berg Cycles
9
10
Duolingo
10
11
Xero Racing
10.5
12
Bakkafrost
13
13
Swisslion Cycling Team
13.5
14
Minions
14
15
Podium Ambition
14.5
16
DK - SVA Zalgiris
15
17
Fablok - Chocolate Jacques
17
18
Kraftwerk Man Machine
18
19
Adastra N Hell Energy Cycling
18
20
Valio - Viking Genetics
18.5
21
GCN Racing
19.5
22
Aramco DP
20
23
Nordstrom - CA Technologies
21.5
24
Azteca - NBCSN
24
25
Campari Asahi Procycling
24.5
26
Andorra Cycling Project
26
After having read all the previews, it's no surprise at all that Mapei easily comes out on top of both tables. Actually, with a median value of 1, this means that more than half of the predictions saw them on top - meaning we have an almost uncontested favorite for promotion.
Furthermore, the community sees Farfetch Pro Cycling in 2nd place, and Repsol - Netflix as well as Desigual promote with both calculation methods as well - with Team Popo4Ever p/b Nemiroff and MOL battling it out for the final promotion spot.
By the way, Mapei and Desigual can be the most confident in those tables, as their predictions vary the least (1st to 8th vs 2nd to 10th).
At the bottom of the table, we have three teams that seem to be doomed - especially Andorra Cycling Project, being dead last in more than half of the previews. But Azteca - NBCSN and Campari Asahi Procycling aren't given a lot of credit, either, both being on a non-relegation spot in just two of the previews each.
And while Nordstrom - CA Technologies and Aramco DP relegate according to both lists as well, they are at least rather close to especially GCN Racing, but Valio - Viking Genetics and Adastra N Hell Energy Cycling don't seem to be very far away, either. Kraftwerk Man Machine have a clear gap in the average list - but quite some part of this is due to the one preview where they end uf winning the division.
Of the newly promoted teams, DK - SVA Zalgiris are given most credit, and they should stay safe from relegation, while the others all seem to be in danger or even doomed. Back-to-back promotions however seem to be excluded this year - it doesn't happen in any of the prediction. Back-to-back relegation doesn't seem to be very probable, either - although Fablok - Chocolate Jacques and Podium Ambition end up doing so in one prediction each.
When speaking of Podium Ambition - they have the second highest "range of inconfidence", with their predictions reaching from 3rd to 25th - which is as much as Xero Racing (4th to 26th). This is only "topped" by Kraftwerk Man Machine (1st to 25th), who can basically land anywhere and someone will be right about them
Unlike for PT, we got some comments on the PCT submissions - and of course we need to share them here!
Adastra N Hell Energy Cycling
Leaders are unproven but definitely quality
So there's pretty much unanimity about first and last, and some agreement about promoting and relegating teams - but anything is possible, as always in this division
Let's go down another level, as we round off the community preview with a look into the managers' CT rankings. We basically have three categories of teams: those relegating from PCT, the returning CT teams, and the brand-new outfits. Will that show in the predictions as well?
Here are the 2020 CT Community Predictions:
Number of predictions taken into account: 8
Prediction by Average rank
Rank
Team
Average prediction
1
Carrefour - ESPN
1.88
2
Philips
2.13
3
Carlsberg - Danske Bank
3.25
4
Centovalli - Fiat
5.38
5
Nordica - Enel
8.00
6
Tryg - Ritter Sport
8.38
7
Sauber Petronas Racing
8.38
8
Team Europcar
8.50
9
BNZ Cycling Team
9.25
10
Guave's Sunlight Cycling
10.38
11
Strava
10.63
12
Project: Africa
11.50
13
Bralirwa - Stevens Bikes
11.75
14
Trans Cycling Team
12.75
15
Los Pollos Hermanos
13.38
16
Eddie Stobart
15.13
17
Glanbia Foods Cycling Project
16.38
18
Sotto il sole di Riccione
16.75
19
Gjensidige Pro Cycling Team
17.00
20
McCormick Pro Cycling
19.25
Prediction by Median rank
Rank
Team
Median prediction
1
Philips
1.5
2
Carrefour - ESPN
2
3
Carlsberg - Danske Bank
3
4
Centovalli - Fiat
5
5
Tryg - Ritter Sport
7
6
Nordica - Enel
8
7
Sauber Petronas Racing
8.5
8
BNZ Cycling Team
9.5
9
Strava
9.5
10
Team Europcar
10
11
Project: Africa
10.5
12
Guave's Sunlight Cycling
11.5
13
Bralirwa - Stevens Bikes
12
14
Trans Cycling Team
12
15
Los Pollos Hermanos
12
16
Glanbia Foods Cycling Project
16
17
Eddie Stobart
16.5
18
Sotto il sole di Riccione
17
19
Gjensidige Pro Cycling Team
18
20
McCormick Pro Cycling
20
According to the available numbers, Philips and Carrefour - ESPN will likely fight it out for 1st and 2nd, with four previews seeing the Dutch outfit on top, and three of them favoring the French. One win vote goes to Carlsberg - Danske Bank.
These three teams are uncontested in terms of promotion, with only the Danes missing out in one single prediction. Centovalli - Fiat seem to be a bit in no-man's land, but should definitely be in contention, too.
When it comes down to the final (direct) promotion spot, however, the Average-based table sees all of Nordica - Enel, Tryg - Ritter Sport, Sauber Petronas Racing and Team Europcar very close together, with BNZ Cycling Team not very far away, either.
The Median-based table has the same teams in 5th to 7th, with just the promoting teams swapped. But then, we see Team Europcar drop from 8th to 10th - a 2nd and 3rd place prediction definitely helped their average, but 4 previews having them just outside the Top 10, they lose some spots. In both calculation methods, they are the lowest ranked relegated team - up to them to prove everyone wrong!
In terms of red lantern, the pretty much unanimous opinion is that McCormick Pro Cycling will finish dead last. 7 out of 8 previews see them at the bottom, with Eddie Stobart taking that not very glorious spot in the remaining preview.
Among the new teams, Nordica - Enel are clearly seen as the best outfit, as they should even have a shot at promotion according to the community! They promote in 3/8 of the predictions, with only Guave's Sunlight Cycling running away with promotion in one preview.
As for PCT, we have a few comments on the CT teams - let's take a look at them!
BNZ Cycling Team
Schultz could be tremendous or terrible at that wage. Good leaders, but all are a bit limited in some way.
Bralirwa - Stevens Bikes
Naesen's great, but not sure how useful on a CT calendar. Options elsewhere, but none stand out.
Carlsberg - Danske Bank
Leader for every terrain, but the standout guy is a sprinter. Also little depth. Good but not as good as the top 2. Prediction: 3rd
Carrefour - ESPN
Way too many sprinters but still top quality everywhere. TTT squad will help for GC and great depth.
Centovalli - Fiat
Moscon + Pernsteiner will be great for PCM 18 hills. Great sprinters too, but held back a little by stage races and cobbles.
Eddie Stobart
A great puncheur squad, but still way too much aging filler taking up cap space.
Gjensidige Pro Cycling Team
The sale of Bystrom leaves a terrific puncheur squad without a true top gun, and they never made up for it.
Glanbia Foods Cycling Project
They should win most TTT's and place in TT GC races, plus Bennett won't be terrible. But lacking too much.
Guave's Sunlight Cycling
Should compete everywhere, but almost always for minor placings.
Los Pollos Hermanos
Baugnies alone will score a lot, beyond that his season depends entirely on Marquez.
McCormick Pro Cycling
Tanking for McNulty. It'll pay off long term, but this year won't be very exciting.
Nordica - Enel
Ratiy will be great and has plenty of support. Two scoring sprinters and decent leaders elsewhere.
Philips
Kelderman is huge obviously, Keinath will score big too. Dumoulin is underrated, and cobbles and sprints are likewise covered.
Project: Africa
Pinot remains top tier, but the TTT squad may be useless if they drop him. Not much to fall back on.
Sauber Petronas Racing
Best stage race team in the division, but other disciplines have been neglected.
Sotto il sole di Riccione
Afraid Dall'Oste will actually be worse as a CT leader. Zaini will have to perform.
Strava
GVA will score, but elsewhere I'm not convinced a puncheur train will be productive in PCM 18, even with their sprint ability.
Team Europcar
Great depth on long or short inclines, but not behind a standout leader. Missing other scorers.
Trans Cycling Team
Not bad and plenty of talents, but despite decent coverage, top points will be hard to come by.
Tryg - Ritter Sport
Lunke + Cort is a great tandem, but maybe a little redundant. Hoelgaard will be great again and other terrains are covered well.
And that's it! Thanks a lot for all of you who participated in this preview, either by doing a full division preview on your own (all of which were taken into account), or by participating by PM. Let's see how well the "common" opinion matches the actual outcomes - and we'll try to track as well who did the best prediction!
Thanks for the previews, very cool to see another one for CT. Not sure why several puncheurs shouldnt work in pcm18 but lets see. A Top 10 would be great.
"It’s a little bit scary when Contador attacks." - Tommy V
Very nice preview. The predictions here seem quite consistent with SN's so that's reassuring. Would be great if we could convert that 11th into a top 10. All depends on Pinot!
I guess 15th is a bit more realistic than the 7th place I got in the mathematical preview which was largely due to no backup stats being taken into account. With some luck I could maybe compete for a top-10, so that'll be my goal
Great stuff, good to see the wisdom of the crowed, Will have to test the performance at the end of the season to see if the consensus preview outperforms the individual ones.
One of your previews I have to questions because if 7 out of 8 had us last and our average is 19.25 I think that means somebody put us 14th. They were clearly in a post-election fog.
I think it'll be a really close race between Carlsberg, Carrefour and ourselves, with me personally putting the order as I mentioned here. A lot depends on how dominant Kelderman can actually be and how well I've planned the others.
I thought that after three months - and right before the GT season which will cause a lot of movement anyway - it's time to take a look at the current standings vs. predictions. For this, I compare the latest rankings with the median prediction, which is rather immune to extreme outliers. Let's start with the PT:
PT Rankings #3 vs. Median Prediction
Team
Current rank
Predicted rank
Difference
Evonik - ELKO
1
1
0
Festina - OAKA
2
7
-5
Moser - Sygic
3
2
+1
eBuddy
4
4
0
ISA - Hexacta
5
20
-15
Isostar - Specialized
6
3
+3
EA Vesuvio
7
11
-4
Team Puma - SAP
8
8
0
Grieg-Maersk
9
16
-7
Aker - MOT
10
10
0
Team Tinkoff - La Datcha
11
13
-2
Bennelong - Mitchelton
12
9
+3
Air France - KLM
13
12
+1
Team UBS
14
14
0
Rakuten Pro Cycling
15
18
-3
King Power
16
17
-1
Generali
17
6
+11
Gazelle
18
5
+13
Aegon - Peroni
19
19
0
Volvo acc. by Spotify
20
15
+5
cycleYorkshire
21
21
0
Indosat Ooredoo
22
22
0
We can see that the median prediction got the current place right for 8 out of 22 teams, which is quite a lot! The average deviation is 3.37 places, but only four teams are outside the +/- 5 places margin:
ISA - Hexacta were obviously hugely underestimated, as they are 15 spots ahead of the predicted place, which is most of any PT team. If we only look at PpRD, they are even 3rd, which is a huge PT debut!
Grieg-Maersk are surely benefitting a bit from the cobbled races, and are likely to be found further down after the GTs - but who knows...
On the other side, Generali and Gazelle, predicted 6th and 5th, definitely didn't have the start they imagined. But with Spilak and Madrazo likely doing a GT and probably taking quite some points from the next few months, they might go exactly in the other direction than Grieg. But surely, they had hoped for a better start to the season, and surely didn't imagine being in the relegation fight by now.
All other teams were predicted +/- 5 spots from where they currently are - including the leaders from Evonik - ELKO and the two teams at the bottom of the table, cycleYorkshire and Indosat Ooredoo.
Let's see how this evolves until the next update, which is likely to be posted by the end of June.
Let's move on to PCT:
PCT Rankings #2 vs. Median Prediction
Team
Current rank
Predicted rank
Difference
Farfetch Pro Cycling
1
2
-1
Repsol - Netflix
2
3
-1
Mapei
3
1
+2
Team Popo4Ever p/b Nemiroff
4
4
0
Valio - Viking Genetics
5
20
-15
Bakkafrost
6
12
-6
Lierse SK - Pizza Ullo PCTeam
7
8
-1
Duolingo
8
10
-2
Andorra Cycling Project
9
26
-17
Polar
10
7
+3
Berg Cycles
11
9
+2
Minions
12
14
-2
Podium Ambition
13
15
-2
Xero Racing
14
11
+3
Desigual
15
5
+10
GCN Racing
16
21
-5
Swisslion Cycling Team
17
13
+4
MOL
18
6
+12
Campari Asahi Procycling
19
25
-6
DK - SVA Zalgiris
20
16
+4
Aramco DP
21
22
-1
Adastra N Hell Energy Cycling
22
19
+3
Fablok - Chocolate Jacques
23
17
+6
Azteca - NBCSN
24
24
0
Kraftwerk Man Machine
25
18
+7
Nordstrom - CA Technologies
26
23
+3
With more teams, it's explicable that the average deviation is higher than in PT, namely 4.5, which is about one place more. However, we can again see that mot teams were predicted within a +/5- places margin - especially the Top 4 correspond with the expectations. We do have a few remarkable outliers, though:
Valio - Viking Genetics surprisingly got almost half their points from sprinter Manninen, who currently is 3rd in the individual standings, which is obviously a huge performance.
The situation is even clearer in the case of Andorra Cycling Project, where Summerhill is dominating the cobbles no matter the race category. It will have to be seen if they can avoid dropping too much once the cobbles season is over.
Going the other way round so far are Desigual, which can at least partly be explained by the underperformance of their leader Tenorio. They're still sitting pretty safe in the middle of the table - but that's by far not enough for a promotion contender.
The same holds for MOL, whose best scorer surprisingly is Waeytens - with their presumed leader Cattaneo clearly underperforming so far.
Finally, at the bottom of the table, we mostly have the teams that were expected to be there, or at least to be involved in the relegation fight.
And last but not least, the CT:
CT Rankings #2 vs. Median Prediction
Team
Current rank
Predicted rank
Difference
Centovalli - Fiat
1
4
-3
Philips
2
1
+1
Carlsberg - Danske Bank
3
3
0
Carrefour - ESPN
4
2
+2
Los Pollos Hermanos
5
15
-10
Tryg - Ritter Sport
6
5
+1
Bralirwa - Stevens Bikes
7
13
-6
Gjensidige Pro Cycling Team
8
19
-11
Guave's Sunlight Cycling
9
12
-3
Project: Africa
10
11
-1
BNZ Cycling Team
11
8
+3
Team Europcar
12
10
+2
McCormick Pro Cycling
13
20
-7
Strava
14
9
+5
Sotto il sole di Riccione
15
18
-3
Eddie Stobart
16
17
-1
Nordica - Enel
17
6
+11
Sauber Petronas Racing
18
7
+11
Trans Cycling Team
19
14
+5
Glanbia Foods Cycling Project
20
16
+4
As in PCT, we have two teams performing a lot better than expected - and two teams doing exactly the other way round. The average deviation is the same with 4.5, meaning it's probably the worst of the three prediction so far.
Again as in PCT, the Top 4, were predicted correctly, with only the order slightly changing, and Tryg - Ritter Sport in 6th place can be considered a good guess as well.
The positive surprises so far definitely are Los Pollos Hermanos and Gjensidige Pro Cycling Team. Whereas the further could have been expected to be there due to 400k cobbles hero Baugnies, the latter have scored quite a lot of surprising points by flat beast Katrasnik. The cobbles will be over some time, but flat stages will always be there - let's see if these teams can stay up there!
Nordica - Enel were predicted to be a promotion contender, and now are towards the bottom of the table. This was always possible, many of the team's riders being rather one-dimensional.
It's however a much bigger surprise to see Sauber Petronas Racing down there as well. However, they have done the least race days of any team so far, with their leader Ranaweera having competed only in Venezuela. They definitely are expected to gain some spots, which their PpRD ratio indicates. But every point first has to be won...
On a positive note, let's mention McCormick Pro Cycling as well, who were predicted to finish dead last, but are doing much better so far. The double K, Keough and Kuss, are working well so far for them - they even have more stage wins than the leaders from Centovalli - Fiat
That's it for the moment. As said before, there will likely be another post pre-TdF, and maybe one just before the closing month.
