the PaddockReport
Projections

Projections

Beta Projections are experimental and still being tuned. Here is what they are, and how they have done so far.

All-time track record Beta

Finals called
55
across 1 show
Favorite won
60%
33 of 55
Winner in our top 3
82%
45 of 55
Champion & reserve in our top 5
76%
42 of 55

By show

Projection track record by show
Show Called Favorite won Winner in top 3 Champion & reserve in top 5
2026 US Arabian Nationals 55 33 (60%) 45 (82%) 42 of 55

How projections work

For a national championship final, once the semi-finals (or qualifiers) before it have posted, we give each horse in the field a chance of winning the championship. The chances in a final add up to 100%.

Locked before the class

Each final is locked once, as soon as its earlier round has posted and before its session starts. That locked forecast is the one we grade, and we never replace it.

The chances can keep moving as results come in until the final's session starts: Locked beside Now, with arrows for the points gained or lost. Those updates are for interest only. They are never graded, and they stop when the final's session starts.

What the numbers mean, and what they don't

  • They are chances, not picks. A 40% favorite still loses more often than it wins.
  • The judges decide. Projections know nothing about how a horse looks or goes on the day, its condition, or a late scratch.
  • They say nothing about a horse's quality beyond its results. They are not advice and come with no promises.

How we keep score

A final counts once its results post, and only if it was locked before the class. A champion we did not project at all counts as a miss.

  • Favorite won: the champion was the horse we gave the best chance.
  • Winner in our top 3: the champion was among our three best chances.
  • Champion & reserve in our top 5: both were among our five best chances.
  • Awaiting results: a final we have locked whose results have not posted yet.

With only a few finals, these numbers can swing a long way on a single result, so we say when the sample is still too small to judge.

Beta

Projections are experimental and still being tuned. When we change how projections work, finals that are already locked keep the forecast they were locked with.