If you spend any time on the various wrestling forums (and I know you do – how else would you find this?) then you have probably spent some time complaining about the various ranking services. And if you are related to a wrestler then you have definitely have spent a lot of time complaining about the various ranking services.

Since all is fair in love, war and wrestling rankings why don’t we turn it around and hand out some rankings to those who would rank?

Meet Our Illustrious Panel

For the 2025-2026 season I kept track of Flo, Intermat, Wrestlestat, and the Coaches’ Rank. These are the only services that I am aware of that rank 33 wrestlers per weight. As a completist we will treat this as the complete list.

Flo subscribes to a results driven process – trying to eliminate opinion or projections from their rankings. I would not say they take a maximalist approach to this, but they come close. As a result they have taken some heat in the past. And will in the future. Flo includes their logic for ranking changes on a weekly basis and maintains the history throughout the season for easy historical review.

Intermat does not publish their philosophy of ranking. But they do call them rankings – not ratings – so we can assume they are of a similar kind to Flo with differences at the margin. Intermat only publishes their most current rankings.

Wrestlestat uses a rating system to produce rankings. Their system is a modified form of the Elo rankings invented by Arpad Elo, a former chess master and physicist, for rating chess players. The system involves a transfer of points between combatants based on the relative rating difference and the result. Wrestlestat rankings/ratings are purely algorithmic. Wrestlestat maintains its history online going back to 2014, making it the most robust historical database.

Coaches’ Rankings are released just a handful of times a year and all after the turn of the year. Sometimes the last set, used for seeding, is released – and sometimes it is not. The roster and methodology have changed over time. But one thing has been consistent, there is a minimum match requirement to be considered eligible to be ranked. Since there is no minimum match requirement to qualify for the NCAA tournament, this can create a significant disconnect. Coaches’ Rankings are, by far, the hardest to find and access in a machine readable format.

Metrics

For these rankings of rankings we will look at a few metrics.

  1. How accurate was the service in predicting the field of 330 wrestlers?
  2. How accurate was the service in predicting the 80 All-Americans?
  3. How accurate was the service in predicting all wrestlers exact finishes?

Before we begin, though we have to acknowledge that this was a particularly tough tournament to predict. With load management becoming more of a thing in wrestling, predicting the field becomes more difficult. This is especially true for the coaches’ rankings where a strict match limit is imposed for consideration. The other services do not have this limitation, but fewer data points makes the job harder on everyone.

The Field

Intermat led the field in predicting the field. Their final rankings contained 306 of the 330 (92.7%) wrestlers who eventually made it to Cleveland. Flo was next with 301 (91.2%), followed by Wrestlestat at 291 (88.2%), and Coaches at 286 (86.7%).

This marks the third year in a row that Intermat has led this category. With 305 correct rankings in 2025 and 312 in 2024, their three year total of 923 leads Flo’s three year total of 903 by almost 7 per year.

The Top Eight

With 79.2% of all points at the NCAA tournament scored by 24.2% of the field – the 80 All-Americans – this might be the most important category. Who did the best job predicting All-Americans? The order flips at the top.

Flo had 55 of the All-Americans among their top 8 ranks (68.8%). Intermat and Wrestlestat were not far behind with 52 each (65%) with Coaches bringing up the rear at just 45 (56.3%).

As mentioned this was a particularly tough year. Even the seeding committee did a poor job of predicting the top 8 wrestlers. Only 50 top 8 seeded wrestlers (62.5%) went on to earn All-American honors. This is the fourth lowest total in history, trailing only 1979, 1980, and 1983. 1979 was the first year with 12 seeds and with 8 All-Americans. Clearly they were working some things out back then with regard to seeds.

Flo also led the rankers in predicting AA’s last year in a very tight race. Flo finished with 55, Intermat with 54, Coaches’ Rank with 53, and Wrestlestat with 51. Flo’s 55 ties the seeding committee’s 2025 performance also.

Now we need to take a minute to appreciate, in hushed, reverant tones, what Flo did in 2024. Three years ago Flo got an absurd 74 of 80 All-Americans correct. At 92.5% they lapped the field. That was in spite of Intermat also putting up their best total of 62 (77.5%). The seeding committee got 60 correct (75%) and Coaches’ Ranking had its best year with 59 correct. Wrestlestat came in with 58 (72.5%) correct answers.1

But here is the really wild stat. Of the 74 Flo got in the top 8, a ridiculous 50 of them were spot on. Fifty times their final rank became the wrestler’s exact finish.

Mind blown.2

The Other Guys

But what about the other guys? The Terry Holtzes and Allen Gambles of the tournament? After all there are are another 250 wrestlers who embraced the struggle. Let us not minimize their achievement. It is time to stand up and count one and all.

To measure overall performance of the ranking services we can look at the sum of absolute differences between the rank/seed and the ultimate finish across the entire field. To do this , we need to make some choices. How do we deal with the fact that the rankers are missing sizeable chunks of the field? And how do we deal with the fact that only the top 8 finish at a specific number? Everyone else finishes within a range.

First, the missing.

I made the decision to assign implied rankings for any wrestlers missing from a ranking service in the order that they were seeded beginning with 34. For example, if the #27, #28, and #29 seeds were missing from a ranking service I assigned them #34, #35, and #36, respectively, for that service. This is the kindest assumption for the ranking services.

Intermat suffered the least from this adjustment with most of the wrestlers they excluded going 0-2. The implied ranking only contributed 2.8% to their total deviation. Flo took a larger hit with 6.6% of their total deviation coming from their non-ranked wrestlers. Wrestlestat and the Coaches really took it on the jaw here with 12.3% and 15.1% of their total deviation coming from this source.

And now, the range.

For wrestlers who finished in a range (Blood Round, Round of 16, Round of 24, 0-2), if their seed was also in the range they finished they were assigned a deviation of zero. Everyone else finishing in a range was assumed to finish at the mid-point of the range. For example, a #9, #10, #11, or #12 seeded wrestler who lost in the blood round would each have a deviation of zero, but a #8 seeded wrestler who lost in the blood round would have a deviation of 2.5 (10.5 minus 8).

With the accounting out of the way we can focus on the results.

Flo just edges out Intermat in this category. With an absolute deviation between their rankings and the final results of 1,561.5, they ease past Intermat’s total of 1,581. Neither could outperform the seeds themselves (especially given the handicap of missing participants) which deviated by “only” 1,483. But Flo and Intermat were both categorically better than Wrestlestat (1707.5) and the Coaches (1812.5).

The 2026 Ranker’s Crown

By taking 2 out of our 3 categories, including the all important All-Americans, the 2026 Ranker’s Crown rests squarely on the heads of the members of the Rankers Guild at FloWrestling. Congratulations to Jon Kozak and JD Rader – great year, keep the rankings coming – but beware.3

With this prestigious award comes a foreboding warning from none other than the famous Renaissance wrestler and All-Stratford Champion of 16044 (who is also said to have dabbled a little in writing ), William Shakespeare, “Uneasy lies the head that wears the crown”.

Don’t get comfortable. Earl Smith is coming for you.

comment below, or
contact me at wrestleknownothing@gmail.com

  1. In the original version of this I mentioned that I had failed to save Wrestlestat data for 2024. Problem solved. Thank you to Wrestlestat for reaching out to me and providing the 2024 data I was previously missing. Very kind. ↩︎
  2. Or I just saved the data incorrectly. Let’s hope not. But it can be an independent study project for anyone ambitious enough to try. ↩︎
  3. But don’t get too big an ego, 2024 JK and JDR are making you look bad. ↩︎
  4. No need to look that up. I am sure it is right. ↩︎