As promised earlier, I have again collected all the predictions out there and merged them into a single "Community Preview". Unfortunately, I only did this for two divisions - PCT and CT - given that only Croatia posted a quantitative PT prediction. Gone are the days of the exhaustive and great-to-read previews by SportingNonsense or SotD it seems
On the other hand, I was able to take into account 7 predictions for PCT and CT each, so plenty of data available - let's see if anyone was smarter alone than the "common wisdom"
Before diving into the analysis, let's recall how I did the final "merged" prediction:
- Find the median predicted rank of every team
- Find the average predicted rank of every team
- Find the range of prediction (i.e. difference between mininum and maximum predicted rank)
- Order first by median, then by average, then by range
The sorting by median values minimizes the effect of outliers. The average is taken as a tie-breaker in case of equal median values. And finally, we'll take the range as an "ultimate" tie-breaker, as I consider a more narrow range to be potentially more precise. However, that final criterion didn't need to be applied, as the first tie-breaker was sufficient in all cases.
Hopefully that was understandable, and now let's move on and take a look at the
2022 PCT Community Prediction
Number of predictions taken into account: 7
Spoiler
These predictions have been taken into account for PCT:
On top, both Amaysim Cervélo and Bralirwa - Stevens Bikes actually get the same average, with the latter being predicted only 1st (2x) and 2nd (5x), while the Aussies have one 4th place prediction. With the median rank being 1 (due to four 1st place predictions), Amaysim are the community's title favorite.
After those two top teams, Carlsberg - Danske Bank is the third team predicted to finish on a promotion spot by all predictions - it looks like they like odd numbers, as they are only predicted 3rd (5x) and 5th (2x).
The race for the remaining promotion spots looks pretty wide open, although Binance and Xero Racing - both with a median 5th place - and 6 Team Popo4Ever p/b Morshynska look like the favorites. However, the latter rather look to be set for 6th place (5x), but at least the Roundtable common prediction sees them on a promotion spot.
Besides those 6 teams, only Philips - Force India (2x 5th), Assa Abloy (1x 4th), Voyagin - Bird (1x 1st) and Red Bull Žalgiris (1x 5th) get at least one prediction on a promotion spot.
In terms of relegation fight, things look to be pretty clear. Project: Africa and Trans Looney Tunes get 100% relegating predictions, while all of Crabbe-CC Chevigny, Sauber Petronas Racing and Strava stay up in one single prediction.
Which means that only three other teams get one relegating prediction each: Indosat Ooredoo (1x 20th), Gjensidige Pro Cycling Team (1x 20th) and HelloFresh - Lampre (1x 23rd). Among those, especially the Indonesians might be heavily at risk, given that nobody predicts them higher than 18th, while Gjensidige and HelloFresh both get some Top15 predictions.
In terms of variance, two teams stick out: Xero Racing, who are predicted to promote in 4/7 previews, also get a 17th place prediction, while Assa Abloy's predictions reach from 4th to 18th. Both are predicted in a range of 14 spots, so they look to be difficult to assess. It has to be said though that for both teams it's mainly one prediction being way off the others.
Let's see how well that community prediction works out this year - so far, it has always been among the more precise previews, even though sometimes an individual one proved to be the best. Stay tuned for CT which will likely come tomorrow!
As you can see in the spoiler, Ulrich Ulriksen and cunego59 submitted two predictions each - both supposing their second one would be the more accurate. I took all of them, so everyone can see which model actually is more accurate
On top of the division, just like in PCT, two teams get only Top5 predictions: Jura GIANTS, who even end up winning the division in 6/7 predictions, and McCormick Pro Cycling, who get a 4th place as their worst prediction. The start of the season clearly puts the latter ahead in terms of title fight; let's see how things move on.
Behind, Tafjord Kraft get 4/7 promoting predictions - with 7th being their lowest value, but three previews even seeing them on the podium. Tryg - Gobyk only get one podium prediction, but they promote in 6/7 cases. So these four teams look like the safest bets for direct promotion spots.
The race for the 5th place is very close according to the community preview, with all of BNZ-Superhero Racing, the newbies from Caja Rural Cycling Team, and BWT Hyundai N Cycling getting 3/7 Top5 predictions. The first and last actually even have two podium predictions, but both get a wider range of potential finish spots than the Spaniards.
Given that the gap especially in terms of average prediction is pretty large to Colombini-Adler Aalterpaint, we can expect the 7 aforementioned teams to fight for the 5 automatic promotion spots. In fact, only Air New Zealand p/b Pirelli and Bianchi - Panerai get one single Top5 prediction each (both taking one 4th place).
On the other end of the table, we see an impressive "0" in the range column - with every single preview predicting the relegating Podium Ambition to finish dead last in their new division. In general, the new teams populate the lower spots in the prediction, with Pas Normal Studios - Mikkeller being the second lowest established team on a predicted 13th place.
We've already seen that there's one outright favorite for the "Newcomer of the Year" title - Caja Rural Cycling Team, who with a good chunk of luck could even achieve direct promotion.
That's it for the moment; let's see how often I'll track the actual standings vs. the predictions. I hope to do so at least twice during the season, and once at the end. But for that, motivation needs to stay up throughout the season - clearly not a given.
Great job mate! Very interesting read, despite PT not being there. And like you say I won't be writing one this season. I was considering it, but in all honesty I thought it wasn't really my turn and wanted to let someone come before me - but as no one seemed to be interested I dropped the plan of writing it. It does take quite some time, and if the interest isn't there - why bother.
I'm really happy to read the CT and PCT previews - although I must admit that it feels a bit silly to not see a single PT preview made by a PT manager. I did very much enjoy the one bbl made though!
Yes, I know that there's only one single (quantitative) prediction for PT, but I still wanted to check how that one compares to reality. And yes, we all know that rankings aren't very meaningful pre-GTs, but let's still do it:
Team
Current rank
Predicted rank
Difference
Isostar - Specialized
1
2
-1
King Power
2
8
-6
EA Vesuvio
3
12
-9
Grieg-Maersk
4
13
-9
ELCO - ABEA
5
5
0
Los Pollos Hermanos
6
21
-15
Tinkoff - La Datcha Team
7
17
-10
ISA - Hexacta
8
16
-8
cycleYorkshire
9
11
-2
Polar
10
18
-8
De Stijl Cycling
11
15
-4
Aker - MOT
12
4
+8
Gazelle
13
6
+7
Huski Chocolate
14
14
0
Aegon - Peroni
15
10
+5
Moser - Sygic
16
7
+9
Team UBS
17
19
-2
MOL Cycling Team
18
9
+9
Evonik - ELKO
19
3
+16
Duolingo
20
20
0
Team Puma - SAP
21
1
+20
Zwift Pro Cycling
22
22
0
One thing is obvious - and easily explicable: the teams focusing more on classics than on stage racing are "green" early on, i.e. they're higher in the rankings than where they're predicted to be at the end of the season. Typical representatives are Grieg, King Power or Tinkoff. Add to those the teams sending strong lineups to early-season stage races like EA Vesuvio, Los Pollos or Polar - and finally notorious overperformers like ISA Hexacta - and you've covered most of the (higher) green numbers.
Obviously, other teams with strong classics specialists had pretty big stinkers, like Aker or Huski. Especially Aker being pretty far down seems a bit odd - or maybe that 4th place prediction is just too high.
On the other hand, we have teams like Puma or Evonik ridiculously far down the rankings - but we all know (or at least strongly suppose) both Herklotz and PHL will do a GT and score 1000+ points on their own. MOL should also be betting on the three-week stage races, as are Moser, Gazelle or Aegon. So that's most of the big red numbers out of the way, too - with just Aker being in a tricky position. However, as they are still 12th and don't really look to be in relegation danger, they probably can live with that - and as said before, maybe that prediction was just a bit optimistic.
On the other hand, the Ardennes are yet to be raced, too - and that's exactly where Aker could make up some spots.
In general, Croatia's prediction currently has an average deviation of 6.7 spots per team - which is not great, but I'm sure it will look far better after the first two GTs, and maybe even better by the end of the year! Thanks for your prediction, mate, I'm still glad to play with, even if it's the only one I have for PT
With 7 predictions, we can indeed speak of a "community preview" for PCT - let's see how this one compares to reality:
Team
Current rank
Predicted rank
Difference
Amaysim Cervélo
1
1
0
Binance
2
4
-2
Lierse SK - Pizza Ullo PCTeam
3
8
-5
Bralirwa - Stevens Bikes
4
2
+2
Carlsberg - Danske Bank
5
3
+2
HelloFresh - Lampre
6
16
-10
Trans Looney Tunes
7
22
-15
Red Bull Žalgiris
8
11
-3
Zara - Irizar
9
12
-3
Sauber Petronas Racing
10
21
-11
Volcanica - Fox
11
17
-6
Cedevita
12
15
-3
Kraftwerk Man Machine
13
14
-1
Xero Racing
14
5
+9
Philips - Force India
15
7
+8
Minions - Subwoolfer
16
13
+3
Voyagin - Bird
17
10
+7
Strava
18
20
-2
Team Popo4Ever p/b Morshynska
19
6
+13
Gjensidige Pro Cycling Team
20
18
+2
Indosat Ooredoo
21
19
+2
Assa Abloy
22
9
+13
Crabbe-CC Chevigny
23
23
0
Project: Africa
24
24
0
We know that RDs spent can still differ by a factor 2 at this time of the year; still, a few tendencies can be observed already. For example, the leader and the bottom 2 are correctly predicted, and according to PpRD, all three teams aren't likely to move a lot.
But the early big surprises, they exist - in both ways. On a positive note, Trans are currently 7th instead of being down on the relegation spots - they're however also in the Top 3 of the teams having raced most so far, and many of those races were on their favorite terrains (hills & cobbles). Projections also see them closer to the relegation than to the promotion zone - let's see if they can prove everyone wrong!
We have a similar picture for Sauber, who are 11 spots above prediction (which would be relegation as well), but they have an even lower projected score than Trans. On the other hand, they've got some huge depth for uphill races - with the lack of TTers, sprinters and top-notch climber it will however still be a lot of work to avoid a drop.
With HelloFresh, we have a third two-digit overperformer - and unlike the others, their PpRD sees them even higher than they currently are! On the other hand, more than half of their points were scored by the duo Daniel - Kennaugh, who obviously don't have enough RDs to carry the team all season long - and not all races will suit the team as much as those they mostly had so far.
But we also had two big losers so far, with the most surprising one probably being Popo4Ever. Instead of having an outside shot at promotion, they're dangerously close to the relegation zone - the lack of mountain stage races so far respectively the underperformance in the one they have already done (Colombia) are surely a factor, but the big points are lacking on all terrains so far. They'll surely be glad once the cobbles are over, though.
And what about Assa Abloy? Predicted to finish 9th, they're currently deep in the relegation zone, and PpRD doesn't give them much hope, either. One reason surely is that their strongest terrain are TTs, but they haven't had the opportunity to play this card yet. They also like mountains - but there weren't many mountains yet, either. And they don't like cobbles - so chances are there they will eventually leave the danger zone behind, but even that is far from granted - let alone claim 9th place as predicted!
Philips, Xero and Voyagin are quite a lot below predictions as well, and at least the two latter will be happy when the cobbles are over. All three clearly are waiting for more uphill races to improve their current rank - let's see if that will eventually happen!
But how precise is the prediction, acutally? Well, after the first three months of the year, it differs by 5.1 spots on average - last year, we had 4.6 at the end of the year, so that's a pretty good first impression.
Many of the individual predictions are very close or even better - Abhi's Roundtable prediction currently provides the best value, with an average 4.8 spots divergence. Croatia's (5.0) and jandal's (5.1) roundtable contribution are also better than or equal to the community prediction, with the final, common Roundtable prediction being slightly worse (5.3). red's 2022 PCT prediction is slightly better with 5.2 - so we have 6 predictions, including the community one, ranging from 4.8 to 5.3 average spots deviation. Let's say that for now the community isn't smarter than some individuals, but let's see how that might change in the next update!
Edited by Fabianski on 08-12-2022 16:11
Preliminary note: Given that Eddie Stobart joined the division after the prediction round, they are not considered (i.e. teams may gain one spot in the "current rank" column compared to reality if Eddie Stobart should be higher than 19th).
So let's take a look at how the community prediction does with the CT teams:
Team
Current rank
Predicted rank
Difference
McCormick Pro Cycling
1
2
-1
Jura GIANTS
2
1
+1
Colombini-Adler Aalterpaint
3
8
-5
Caja Rural Cycling Team
4
6
-2
BWT Hyundai N Cycling
5
7
-2
Glanbia
6
10
-4
Bianchi - Panerai
7
14
-7
BNZ-Superhero Racing
8
5
+3
Tryg - Gobyk
9
4
+5
Tafjord Kraft
10
3
+7
Oktal Pharma - TotalTV
11
11
0
Staffbase - BHP
12
17
-5
Bonduelle - Eesti Energia
13
15
-2
Pas Normal Studios - Mikkeller
14
12
+2
Air New Zealand p/b Pirelli
15
9
+6
Bordeaux Métropole - Euskotren
16
16
0
Le Creuset
17
13
+4
Podium Ambition
18
18
0
One thing jumps out: there are no two-digit divergences in CT - probably confirming its status of "easiest to predict division". In general, relegating teams are predicted to bounce back up, while new teams usually take the lowest positions. But hey, Bianchi proves them all wrong for now! They've been doing a good job so far, especially in the hills and on the cobbles, and a result are up in 7th place instead of 14th. Projections see them at the lower end of the Top 10, but they could still end up being one of the positive surprises of the year!
The same holds for Colombini, who are 5 spots above prediction - mainly due to a very strong Saber, but also Vliegen has been flying this year. It's still a long way to go, but they definitely are the early favorites for "best new team" - and even have a shot at first-year promotion! Let's see if they can keep up their pace for the remaining 7 months...
Staffbase also are 5 spots above prediction - which although lower in the rankings is still as respectable. They will be heavily relying on their sprinters and puncheur depth - but especially in those regions of the rankings a great result here and there can make a big difference!
And I also want to mention (once again) Glanbia - predicted to finish 10th, they're currently 6th, and most of the (T)TT-heavy races are yet to come! Of course they heavily depend on Tachira winner Talansky so far, but personally I wouldn't be surprised to see them even among the Top 5 by the end of the year.
But we also do have some rather big disappointments so far. The biggest one surely are Tafjord, who were seen as a safe bet for promotion (3rd), but currently are 10th. Their season start in February surely didn't help for early scoring, but even then their star puncheurs didn't perform as required to be high up in the rankings. Breen is having a decent cobbles campaign, but they will need sprinters and climbers to jump in as well if they really want to achieve promotion!
The situation is quite similar for Tryg, who were predicted in 4th and right now are 9th, mainly due to some fluky performances by star rider Schlegel. However, Kanter hasn't raced a lot, and Hoelgaard has been doing well so far - let's see if they can turn their projected points into actual ones and indeed finish 4th like their PpRD suggests.
Air New Zealand are even 6 spots lower than the prediction -which saw them inside the Top 10. Although there's quite a gap in PpRD between 10th and 11th currently, they're indeed the team that could get closest, having the 11th PpRD average and not having raced a lot yet. They'll definitely need their climbers to perform to move up, though - and Scully to conserve his Guadeloupe form and continue to score.
All in all, the community prediction is already pretty accurate with a 3.1 spots deviation per team. However, Ulrich Ulriksen's CT predictions confirm that he's become an expert of the division, with his first approach being 2.8 spots off - and the second one just 2.1! An incredible value so far, congrats!
But let's also mention cunego's second prediction, which differs by just 2.7 spots per team and hence is the second-best CT prediction out there currently! And let's also mention the Continental Division Songfestival, whose outcome is almost as good as the community preview with an average deviation of 3.2 spots per team.
That's it for now - obviously a lot can still change, but for now it definitely looks like CT is easiest to predict. There's also the least amount of teams (18 vs. 22 vs. 24), but especially Ulrich's predictions are really great!
Thanks for the update on these! In CT it's safe to say that both Tafjord and Tryg haven't had the start to the season that the experts predicted going to be interesting to see if they bounce back since they have very strong teams on paper. Maybe Schlegel will challenge Altur for worst training spend in MG history? Only time will tell!
For PCT and us personally I still think that predicted us 16th was more than fair, with quite a few race days used up by our primary captains in their target races already a race like Baltic in the coming week will be super important for us when it comes to figuring out if we can upset and finish in the top 10. Assa Abloy and Popo not having the starts that people predicted, it is going to be an interesting couple of months to see if those two teams can bounce back and ensure that they reach their season goals, especially Assa Abloy has a super interesting project in the works granded that they survive, which they have enough quality to ensure they do.
Thanks for the write-ups. I guess rankings are a little too early to tell the real picture in any division with it being focused on specific terrains so far, but we'll see. Really looking forward to the next months!
whitejersey wrote:
Maybe Schlegel will challenge Altur for worst training spend in MG history? Only time will tell!
If there is only one good thing to come out of the previous two seasons of T3A PCT struggles, it's establishing this Altur record as fact! (ignore his solid PT result this season please )
I feel like lack of Cobblers is a bit harsh on us this season considering Tom David is currently sat top 20 in the PCT Individual rankings and we have 4 domestiques for him that are similarly focussed in hilly cobbles.
sammyt93 wrote:
I feel like lack of Cobblers is a bit harsh on us this season considering Tom David is currently sat top 20 in the PCT Individual rankings and we have 4 domestiques for him that are similarly focussed in hilly cobbles.
I guess I shouldn't do this kind of writeups late in the evening when I'm still half ill
Of course that should read "lack of sprinters", will fix this one. On the contrary, David clearly is one of the main reasons why you're way higher up than predicted so far! But I stand by the rest - no TTer, no top climber (given this year's climber field) Although I have to admit that Pomoshnikov and Frankiny hurt us more than once last year by beating "top climber" Quintana
Sorry for not having posted more intermediate results, but I was just too busy with other MG and RL things. Nonetheless, here come the final results!
Final PT Rankings vs. "Community" Prediction
Sadly, there was no real community prediction for the highest division, as only Croatia dared posting a preview including ranking predictions. Let's see how accurate that one turned out to be:
Team
Final
Predicted
Difference
Isostar - Specialized
1
2
-1
King Power
2
8
-6
Team Puma - SAP
3
1
+2
Aker - MOT
4
4
0
Tinkoff - La Datcha Team
5
17
-12
MOL Cycling Team
6
9
-3
ELCO - ABEA
7
5
+2
ISA - Hexacta
8
16
-8
EA Vesuvio
9
12
-3
Gazelle
10
6
+4
Grieg-Maersk
11
13
-2
Polar
12
18
-6
Evonik - ELKO
13
3
+10
Moser - Sygic
14
7
+7
cycleYorkshire
15
11
+4
De Stijl Cycling
16
15
+1
Huski Chocolate
17
14
+3
Duolingo
18
20
-2
Aegon - Peroni
19
10
+9
Team UBS
20
19
+1
Los Pollos Hermanos
21
21
0
Zwift Pro Cycling
22
22
0
Croatia apparently was too modest - or just underestimated his team - as he predicted Puma - SAP to win it all. The Germans did well indeed, but finished "only" 3rd - while Isostar easily completed the three-peat. 2/3 podium spots isn't bad - although pretty much anyone would probably have seem these two up there.
King Power though were quite a lot better than expected, taking the runner-up spot inestead of "just" finishing in the lower Top 10. What I find impressive, though, is the spot-on 4th place prediction of Aker - it's rare to see a newly promoted team that high up, and probably even more rare to see this coming!
However, as great as the Aker prediction was, as far off was the one for Tinkoff. Predicted to only just reach safety, they actually did a great job to end up 5th! Another team that finished clearly higher than expected - although with "only" an 8 spots difference - were ISA - Hexacta. Polar taking 12th place instead of relegating are the last team where the final outcome was more than 5 spots higher than predicted.
There were also three teams that finished clearly worse than expected by Croatia. First and foremost, Evonik, who were predicted to end up on the podium - but in the end even missed the Top 10. While the consequences aren't that bad for the Latvians - except for a failed goal - Aegon finishing 9 spots below the prediction had the worst possible consequences: relegation. Moser were the 3rd team more than 5 spots worse than expected, finishing 14th instead of 7th.
Speaking of relegation, our sole predictor had a pretty good feeling there, getting 4/5 right. Duolingo, UBS, Los Pollos and Zwift were all expected to go down - the two last names are actually the only other precise predictions besides Aker. As said before, Aegon was definitely never expected to go down; the prediction saw Polar take the final relegating spot.
Overall, the prediction has a deviation of 3.9 spots on average - which as expected is far better than the value from the intermediate standings (6.7).
Unfortunately, we can't compare this value to any other predictions - I'm hoping to get some more for 2023 again (especially given that of one this year's PCT predictors just got promoted )!
We had 7 predictions for PCT (which you can find in the first post); thereof Abhi's Roundtable contribution was the most accurate early in the season. Let's see how that changed now - here are the final standings compared to the community prediction:
Team
Final
Predicted
Difference
Amaysim Cervélo
1
1
0
Zara - Irizar
2
12
-10
Minions - Subwoolfer
3
13
-10
Cedevita
4
15
-11
Carlsberg - Danske Bank
5
3
+2
Binance
6
4
+2
Kraftwerk Man Machine
7
14
-7
Philips - Force India
8
7
+1
HelloFresh - Lampre
9
16
-7
Lierse SK - Pizza Ullo PCTeam
10
8
+2
Bralirwa - Stevens Bikes
11
2
+9
Assa Abloy
12
9
+3
Xero Racing
13
5
+8
Team Popo4Ever p/b Morshynska
14
6
+8
Red Bull Žalgiris
15
11
+4
Sauber Petronas Racing
16
21
-5
Trans Looney Tunes
17
22
-5
Indosat Ooredoo
18
19
-1
Voyagin - Bird
19
10
+9
Gjensidige Pro Cycling Team
20
18
+2
Strava
21
20
+1
Volcanica - Fox
22
17
+5
Crabbe-CC Chevigny
23
23
0
Project: Africa
24
24
0
Just like in PT, there were only three exact predictions - here we have the winner and the bottom 2. Amaysim were expected to win the division in 4/7 previews, with 2/3 of the others seeing them finish 2nd (and the final one at least had them promoting as well). So that was a rather easy one.
However, three of the other 4 promoting teams were completely unexpected - as all of them were seen to even miss out on the Top 10! Zara and Minions at least got two Top 10 predictions each (though no promotion spot), whereas Cedevita's highest predicted finish was 14th! So yes, PCT sometimes yields some really, really big surprises!
Carlsberg, on the other hand, were predicted to promote in every single preview, although they "only" finished 5th instead of 3rd. They'll get over it, I'm sure.
Binance also finish just two spots lower than predicted - but for them the outcome is worse, as instead of promoting they just miss out - unless there's at least one disband standing once the continuation deadline is over. Or if MGUCI decides to change the rules and abandon "disband promotions".
We've already seen the three biggest surprises - which also are the only two-digit deviations compared to the prediction. The only others finishing more than 5 spots higher than expected are Kraftwerk (but who can be surprised by them doing well anyway?) and HelloFresh, who were able to build up upon their great 2021 CT season and be very competitive in PCT.
But if there are some massive overperformers, the opposite must be true as well. Indeed, we have four teams doing far worse than expected. Most notably Bralirwa, who were predicted to finish 1st or 2nd in every single preview (having the same average prediction as Amaysim!), but in the end don't even finish inside the Top 10! Madrazo obviously was the key factor for this massive disappointment, maybe alongside Stallaert not being quite as dominant as last year.
Xero and Popo4Ever are pretty much in the same boat; they were expected to finish 4th and 5th, but in fact ended up 13th and 14th. Whereas the former had great leaders but not much depth (plus riders like Stannard not performing at all), the latter often had quite a lot of bad luck, losing races by a couple of seconds only. No promotion for both of them.
The other team (besides Bralirwa) ending up 9 spots below expectation are Voyagin - who only just escape relegation! They even got one win prediction, plus three more Top 10s - but both Kinoshita and Shikai didn't perform well enough to earn the team a Top 15 finish.
In terms of relegating teams, the community prediction has 3/5, with P:A and CCC expected to finish on the final two spots, and Strava also predicted to go down. Volcanica and Gjensidige were expected to be in the relegation fight, but rather be successful (Volcanica by all predictors, Gjensidige by all but one) - unlike Sauber and Trans, who were predicted on the bottom spots (Trans by all, Sauber by all but one).
So, what does that mean in terms of numbers? The community prediction has an average deviation of 4.7 spots per team. A value that's matched by one individual prediction - and beaten by four! So common wisdom clearly wasn't better than individual one here.
The winner of the Most accurate 2022 PCT Prediction award is... jandal with his Roundtable Analysis contribution! He missed the final outcome by 4.4 spots per team on average - so he wasn't far better than the community prediction, but he was better! Congratulations!
In 2nd place, we have two contributions: On one hand, Croatia's Roundtable Analysis contribution, and on the other hand the Roundtable Analysis itself! They are 4.5 spots off on average.
Again only slightly worse is redordead's 2022 PCT prediction, with an average deviation of 4.6. So at least the conglomerate didn't mop up the Top 5 spots...
Next are the community prediction itself, matched by knockout's contribution to the Roundtable Analysis. As said previously, both are 4.7 spots off per team. But all of these are really close together.
The remaining two predictions are slightly further off the benchmark, with early leader Abhi's Roundtable Analysis contribution turning out to be less exact at the end of the year than early on (5.3 spots off), and TinxiaS' post in the 2022 predictions for lazy/busy wizards with an average deviation of 6.9.
Thanks to you all for contributing, it's always interesting to see how the predictions compare to reality by the end of the year
![imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg](https://imgur.com/KlJUqDx.jpg[/img[img]]https://imgur.com/yUygrQ.jpg)
