Patience is fine -- but it's more like 2-3 months versus 2-3 weeks. I think some work could be done here, but it seems like a bunch more have been coming in the last few days.

I did Oregon & I'm really surprised at how low the scores are across the board. Big Metz won the race & got a 116. Some winners of AG only races have gotten the same &, while those AGers are really really good, I don't think they would have taken it to him & a lot of other pros there that day. Only 2 elite scores from age groupers in what felt like a really deep race. Seems really low compared to other races.

It seems like you get super inflated scores on warmer days when more people in the middle of the race might struggle with pacing. But that shouldn't erase good performances on good weather days. It's harder and harder to make improvements as race times come down. I don't think a good weather day should automatically mean lower scores up front, and that's certainly what it seems like.

Bumping this up again. Did AG Nats & got a higher score in the Olympic than the Sprint even though I thought I had a much better race on day 2. I know it's about how others perform so at a very basic level does that mean the average athlete had a good race? Do they factor in course length into the ratings? Like they had the bike course listed as 10k but it was longer. Does that matter when scores are compiled?

Just looks like, even for the winners, score are down ~3 points from day 1 to day 2. I just don't get it. Was the winner really that much worse from one day to the next? Seems like a decent enough of a jump that I figured I'd ask.

Yes, the average performance (not the middle of the pack finishing time, but the middle of the pack in terms of performance relative to past performance) on Sunday was a "better day" and on Saturday the average performance was a "worse day."

Conditions on Saturday were poor (hot and humid) so many people performed relatively poorly. Those that handled the conditions better (or started and finished earlier) and performed close to their past performances will likely see a higher score than they are used to (this is the case for me).

Sunday's conditions were pretty good and many people probably performed as expected so to get a higher score than you're used to you'd have to perform exceptionally well.

In short, a "good day" Saturday when the average performance was a "poor day" is worth the same as a "great day" Sunday when the average performance was an "average day."

Course length is not a factor in the ratings.

So I'm super nerdy and I calculated the PAR time for the Olympic Distance Nationals by hand (excel) for a couple of age groups for Saturday's race. I wanted to see if there was a huge difference between the waves that started out at 7 am versus those like mine (M35-39) that started after 9am. The run (heat/humidity) for my age group was pretty brutal in my opinion and I certainly ran slower than expected despite being in pretty good run fitness. What I found was:

M25-29: PAR 2:07:58 / 7:22 am
M30-34: PAR 2:07:47 / 7:12 am
M35-39: PAR 2:11:24 / 9:05 am
M40-44: PAR 2:09:57 / 9:35 am

The race's actual PAR came out to be 2:10:53 (5 minutes slower than the PAR in 2021). There was a over 3 minute difference between PAR times for the waves that started 2 hours later. To their credit, M40-44 performed better than my M35-39 age group despite starting even later. Certainly competitors in the waves after 9am would have finished significantly higher in the overall results had everybody started at the same time. It certainly would not have had any impact on the overall national champion seeing as Matthew Guenter finished over 8.5 minutes ahead of the top finisher in the 9:00am and later waves, but in future years it certainly could if conditions like this year were encountered. My calculations are still a small subset of the overall race but I figured these would be good age groups to compare as there would be less likelihood of drastic year over year changes in ability. Anecdotally, this seems right to me but it's also more or less what I expected to see. Personally I had my highest USAT score ever in this race despite being 3 minutes slower than last year (granted the swim and bike courses were longer/slower this year) and certainly not being in my all-time peak fitness.

How this relates to Sunday's race and the resulting scores is that all competitors started much closer to the start of the race. It didn't get significantly hotter as the final competitors hit the course. As a result, late starters didn't drag the PAR score slower in comparison to what happened on Saturday. As a result, the winner on Saturday scored significantly higher than the winner on Sunday (despite being the same person).

Thanks -- that's super helpful. I think I'm starting to understand it. It still seems like you get penalized for doing well on decent weather days. Shouldn't the rankings be more/less consistent with athlete fitness at a given time? It's just weird to be 2-3 points different from 1 day to the next.

The Obstri rankings seem to be a little more balanced. I'm from the northeast and traveled for a summer 70.3. USAT scores ended up being low at my race. The weather was pretty good. Scores were high at a local 70.3 that had decent but not as good weather. Obstri has me ranked above people at Maine that USAT doesn't. If I had gone to the local race, I would have scored higher. So we're not getting a true snapshot of athlete fitness in real time. We're saying athlete A is better than athlete B because they did a race with more drop-off in the middle of the race.

Interesting comparison with those age groups! I've said it in other threads but I'm in favor of lining everyone up together if possible. People who started later would have done better had they started earlier.

The race's actual PAR came out to be 2:10:53 (5 minutes slower than the PAR in 2021)//

This comparison along with any others from other races, has no bearing on what you did on that day or ranking points. It can be an interesting number when comparing the actual lengths of the course's with weather conditions, but that would be all.


And interesting that the later wave of the 40+ did basically as well as the earlier younger waves. Perhaps you all in the later waves got some wind advantage that the earlier waves didnt?

I wouldn't look at it as getting penalized in the Sunday situation. Rather, you were rewarded for a better result relative to others on Saturday (which likely at least partially aided by your start time).

We're in the same boat. I scored higher on Saturday than Sunday (by quite a bit, partially due to some race execution mistakes on Sunday), but I don't view it as being penalized. As we can see from the data posted above, the ratings are likely a little inflated for early starters Saturday due to the differing start times. This is unfortunate and a weakness of the system for sure. For me, my outlier race score is Saturday's compared to my other scores this year. I nailed my execution and got lucky with my start time. Sunday's score was more in-line with my other scores. I feel I was rewarded Saturday, but not penalized Sunday.

And I guess I'd argue that the ratings should reflect race performance, not fitness. Fitness I suppose can have a pretty broad definition, but ultimately triathlon race performance is composed of physiological fitness as well as a host of other aspects like nutrition planning and execution, pacing strategy, environmental condition mitigation, psychological state, etc. So it seems entirely reasonable to me to have different scores on consecutive days. Also, those that raced Saturday and Sunday were racing with significant fatigue Sunday, while some people just raced Sunday, so it makes sense that we might not have as good of a performance Sunday.

Thanks for doing that analysis! It is about what I would have expected. Seems like the race was 2-3 minutes easier for earlier starters.

The age groups were pretty large for this race, but there's going to be a good bit of noise in the PAR times when using small groups of athletes. This happens at very small local races too where ratings can be funky due to the PAR time coming from a relatively small number of results. I wouldn't read much into the 35-39 vs. 40-44 difference than that.

No I get that I just got my lowest score rating of the year from Sunday & was definitely in better shape than early season races & definitely executed better but because the middle of the race did relatively well I somehow was worse compared to my early season races. My highest rating on the year comes from a Sprint I did on a horrible weather day a few weeks before my goal 70.3. I went in with no taper/with really high volume. I peaked for the 70.3 a few weeks later & crushed it (fast time, good weather day) but scored lower than the Sprint. Should I have peaked for the Sprint because I could have done even better and scored even higher than what I did? I was 1 point off of the elite standard at the Sprint but 4 off at the 70.3. I beat some pros at the 70.3 & know I would have been closer to them at the 70.3 than the Sprint if any had been in that race. I know I performed better at the 70.3 but I didn't perform better compared to the masses because the weather was decent. You get a higher score when others fall off versus when you have a good race. It's making me rethink what races should be important.

I do think it's kind of odd that you can't look at two 70.3's and know who had more fitness on the day. At 70.3 Maine, the top-39 AGers scored 100+. At Oregon, the top-23 scored 100+. Are you telling me nobody from Oregon 24-39 beats anybody from Maine if lined up on the same day? Obstri doesn't seem to think so. The way I'm understanding the rankings is that a 106 from Maine could be a 102 in Oregon. But then why not have a system where those numbers are the same because the athlete is getting the score based on one of them?

I wouldn't let it make you rethink what races are important but I would rethink how important the rankings are. There can certainly be a range of a few points in any direction on any day just based on how things played out. It rare to see days like StG where there's almost a 15 point swing but they do happen. 2 to 3 point swings seem to happen fairly regularly to me. Where you end up in the rankings at the end of the year does not matter. The guy one spot higher or lower than you isn't automatically faster or slower than you. What matters is beating people in the races. I had my highest USAT score on Saturday at 104.2, but I know that it wasn't my best race ever. The numbers are interesting, but nothing more than that.

Now if you're trying to get elite qualifying scores to earn a pro card that becomes a little different. Granted you're limited to races with 500+ people for the score to actually count and there's only a handful of short course races that meet that criteria so you almost have to do middle distance or long course. Think of all the people that got an elite qualifying score at StG that would typically score in the 90s. If you find another race like that within a year you're a pro that does a 2:15 olympic, lol. I'm not too far off from earning a pro card based on the points criteria but I'm laughably far away from any of the other criteria and would be mopped up by any real pro. Perhaps they should drop this qualification criteria or substantially change the rankings system if they want to use it.

I would not try to tell you that the system is perfect as it is, but it's not like the system is designed to give the variation you're pointing out. I just think it would be impossible to design such a system, and even if one were to drop from the sky, I have no idea how one would test it to check if it was indeed perfect.

If you have a score related goal, i.e. getting a pro card, it's probably worth game-planning which races to prioritize. Otherwise, it's just a curiosity. I have some races locally where the courses are a bit tougher and there are usually some higher scores on offer (high 90s to low 100s). Some shorter flatter courses it seems no one is scoring higher than a 93 or so. I don't think there are obvious fixes to that.

yeah, I think you are being unrealistic about the scoring system and how potentially accurate it can be.

Your score for the race is calculated roughly off of how all the other racers do. If you enter a race where 100% of the racers who count in the scoring system have an A++ day and crush their PRs, you'll score a lot lower than if you had entered a casual sprint race where a good number of the racers either not tapering since it's an A race for them, and also probably attracts a lot more casual racers who aren't die hard about the sport as would a world championship event.

In that regards, I do agree with you that you can outperform in races where you happen to manage the tough environmental factor better than everyone else. If you're used to training in summer AZ heat, you'll wilt a lot less in an egregiously hot race, and likely outperform. Simlarly, if you are used to training in rain and sketchy road surfaces you'll outperform in rainy messy races for your scoring.

I also don't get your conclusions re 70.3 Maine vs Oregon. There were more AGers 100+ in Maine but that doesn't mean nobody from oregon can beat the Maine people. It likely means that more historically very fast people (AG100ish) showed up to Maine vs Oregon, but doesn't mean that nobody from Oregon can beat them. I do think you can conclude from that though is that you'll have a slightly tougher time winning the AG or podium in Maine given the larger number of 100 scoring racers there.