Every graded pick this site holds, the back-tested years and the live season together. 1,953 graded picks from 14 January 2023 to 10 October 2026, of which 478 were published as toss-ups and are counted like any other pick.
| Bucket | n | Mean predicted | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| 50–55% | 476 | 52.5% | 50.8% | 46.4–55.3% | −1.7 |
| 55–60% | 456 | 57.4% | 63.2% | 58.6–67.5% | +5.8 |
| 60–65% | 360 | 62.5% | 70.8% | 65.9–75.3% | +8.4 |
| 65–70% | 276 | 67.3% | 75.7% | 70.3–80.4% | +8.4 |
| 70–75% | 189 | 72.4% | 81.5% | 75.3–86.4% | +9.1 |
| 75–80% | 116 | 77.2% | 77.6% | 69.2–84.2% | +0.4 |
| 80–85% | 66 | 82.2% | 86.4% | 76.1–92.7% | +4.2 |
| 85–90% | 14 | 86.8% | 85.7% | 60.1–96.0% | −1.0 |
| All buckets | 1,953 | 62.2% | 66.9% | 64.8–69.0% | +4.7 |
| Sportsbook bucket | n | Mean implied | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| under 50% | 153 | 43.7% | 47.1% | 39.3–54.9% | +3.3 |
| 50–55% | 592 | 50.6% | 60.0% | 56.0–63.8% | +9.4 |
| 55–60% | 201 | 57.6% | 59.2% | 52.3–65.8% | +1.6 |
| 60–65% | 238 | 62.7% | 71.8% | 65.8–77.2% | +9.2 |
| 65–70% | 236 | 67.5% | 66.9% | 60.7–72.6% | −0.6 |
| 70–75% | 184 | 72.3% | 76.6% | 70.0–82.2% | +4.3 |
| 75–80% | 175 | 77.3% | 80.6% | 74.1–85.8% | +3.3 |
| 80–85% | 96 | 82.0% | 81.2% | 72.3–87.8% | −0.8 |
| 85–90% | 61 | 87.3% | 93.4% | 84.3–97.4% | +6.1 |
| 90–95% | 16 | 92.2% | 87.5% | 64.0–96.5% | −4.7 |
| 95–100% | 1 | 95.4% | 100.0% | 20.7–100.0% | +4.6 |
| All buckets | 1,953 | 61.8% | 66.9% | 64.8–69.0% | +5.1 |
The sportsbook series bins the same picks by the sportsbook’s own probability for the side we picked. Where that is under 50% the sportsbook had our pick as the underdog; those rows are the “under 50” bucket rather than being dropped.
Until 16 September 2026 the market price we served could come from a prediction exchange as well as a sportsbook, so some 2026 rows in this series carry an exchange price; where our market number came from, before 2026-09-16 sets out how many and what we cannot establish.
Why no bucket sits under 50%. A pick is always the favoured side, so the lowest number FightIQ can publish on a pick is 50% and the lowest bucket is the coin-flip one; the sportsbook series can sit under 50% because it is the sportsbook’s view of the side we picked. One caveat on the combined view. The model behind these numbers changed across the years, so the combined line is the calibration of what we published each season, not of one fixed method.
A back-tested year: the numbers were generated on the season's fights after the fact and scored against the results, which is a weaker claim than a live pick and is labelled as one. 504 graded picks from 14 January 2023 to 16 December 2023, of which 131 were published as toss-ups and are counted like any other pick.
| Bucket | n | Mean predicted | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| 50–55% | 130 | 52.6% | 51.5% | 43.0–60.0% | −1.0 |
| 55–60% | 120 | 57.3% | 60.8% | 51.9–69.1% | +3.5 |
| 60–65% | 99 | 62.5% | 75.8% | 66.5–83.1% | +13.2 |
| 65–70% | 70 | 67.3% | 78.6% | 67.6–86.6% | +11.2 |
| 70–75% | 45 | 72.2% | 73.3% | 59.0–84.0% | +1.2 |
| 75–80% | 27 | 77.1% | 85.2% | 67.5–94.1% | +8.1 |
| 80–85% | 11 | 81.9% | 90.9% | 62.3–98.4% | +9.0 |
| 85–90% | 2 | 85.7% | 100.0% | 34.2–100.0% | +14.3 |
| All buckets | 504 | 61.5% | 67.1% | 62.8–71.0% | +5.5 |
| Sportsbook bucket | n | Mean implied | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| under 50% | 53 | 42.7% | 49.1% | 36.1–62.1% | +6.3 |
| 50–55% | 128 | 50.9% | 56.2% | 47.6–64.5% | +5.4 |
| 55–60% | 63 | 57.5% | 58.7% | 46.4–70.0% | +1.2 |
| 60–65% | 65 | 62.8% | 78.5% | 67.0–86.7% | +15.7 |
| 65–70% | 67 | 67.4% | 67.2% | 55.3–77.2% | −0.2 |
| 70–75% | 46 | 72.4% | 73.9% | 59.7–84.4% | +1.5 |
| 75–80% | 41 | 76.9% | 92.7% | 80.6–97.5% | +15.8 |
| 80–85% | 27 | 81.6% | 77.8% | 59.2–89.4% | −3.8 |
| 85–90% | 14 | 87.3% | 100.0% | 78.5–100.0% | +12.7 |
| All buckets | 504 | 61.3% | 67.1% | 62.8–71.0% | +5.7 |
The sportsbook series bins the same picks by the sportsbook’s own probability for the side we picked. Where that is under 50% the sportsbook had our pick as the underdog; those rows are the “under 50” bucket rather than being dropped.
A back-tested year: the numbers were generated on the season's fights after the fact and scored against the results, which is a weaker claim than a live pick and is labelled as one. 513 graded picks from 13 January 2024 to 14 December 2024, of which 127 were published as toss-ups and are counted like any other pick.
| Bucket | n | Mean predicted | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| 50–55% | 127 | 52.7% | 54.3% | 45.7–62.7% | +1.6 |
| 55–60% | 133 | 57.3% | 62.4% | 53.9–70.2% | +5.1 |
| 60–65% | 93 | 62.5% | 67.7% | 57.7–76.4% | +5.3 |
| 65–70% | 77 | 67.3% | 72.7% | 61.9–81.4% | +5.4 |
| 70–75% | 44 | 72.4% | 86.4% | 73.3–93.6% | +14.0 |
| 75–80% | 28 | 77.1% | 75.0% | 56.6–87.3% | −2.1 |
| 80–85% | 10 | 81.9% | 90.0% | 59.6–98.2% | +8.1 |
| 85–90% | 1 | 86.4% | 100.0% | 20.7–100.0% | +13.6 |
| All buckets | 513 | 61.5% | 66.3% | 62.1–70.2% | +4.8 |
| Sportsbook bucket | n | Mean implied | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| under 50% | 38 | 44.1% | 36.8% | 23.4–52.7% | −7.3 |
| 50–55% | 173 | 50.4% | 63.6% | 56.2–70.4% | +13.2 |
| 55–60% | 50 | 57.7% | 56.0% | 42.3–68.8% | −1.7 |
| 60–65% | 50 | 62.5% | 70.0% | 56.2–80.9% | +7.5 |
| 65–70% | 69 | 67.5% | 66.7% | 54.9–76.6% | −0.9 |
| 70–75% | 51 | 72.1% | 76.5% | 63.2–86.0% | +4.4 |
| 75–80% | 41 | 77.2% | 80.5% | 66.0–89.8% | +3.3 |
| 80–85% | 19 | 81.5% | 78.9% | 56.7–91.5% | −2.6 |
| 85–90% | 20 | 87.5% | 90.0% | 69.9–97.2% | +2.5 |
| 90–95% | 2 | 91.2% | 100.0% | 34.2–100.0% | +8.8 |
| All buckets | 513 | 61.2% | 66.3% | 62.1–70.2% | +5.1 |
The sportsbook series bins the same picks by the sportsbook’s own probability for the side we picked. Where that is under 50% the sportsbook had our pick as the underdog; those rows are the “under 50” bucket rather than being dropped.
A back-tested year: the numbers were generated on the season's fights after the fact and scored against the results, which is a weaker claim than a live pick and is labelled as one. 515 graded picks from 11 January 2025 to 13 December 2025, of which 135 were published as toss-ups and are counted like any other pick.
| Bucket | n | Mean predicted | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| 50–55% | 135 | 52.4% | 47.4% | 39.2–55.8% | −5.0 |
| 55–60% | 129 | 57.5% | 66.7% | 58.2–74.2% | +9.2 |
| 60–65% | 105 | 62.4% | 67.6% | 58.2–75.8% | +5.2 |
| 65–70% | 73 | 67.3% | 72.6% | 61.4–81.5% | +5.3 |
| 70–75% | 38 | 72.4% | 84.2% | 69.6–92.6% | +11.8 |
| 75–80% | 21 | 77.2% | 90.5% | 71.1–97.3% | +13.2 |
| 80–85% | 11 | 82.3% | 81.8% | 52.3–94.9% | −0.5 |
| 85–90% | 3 | 85.8% | 100.0% | 43.9–100.0% | +14.2 |
| All buckets | 515 | 61.2% | 65.4% | 61.2–69.4% | +4.3 |
| Sportsbook bucket | n | Mean implied | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| under 50% | 30 | 44.3% | 63.3% | 45.5–78.1% | +19.0 |
| 50–55% | 211 | 50.3% | 58.8% | 52.0–65.2% | +8.4 |
| 55–60% | 47 | 57.5% | 57.4% | 43.3–70.5% | 0.0 |
| 60–65% | 61 | 62.5% | 75.4% | 63.3–84.5% | +12.9 |
| 65–70% | 54 | 67.6% | 68.5% | 55.3–79.3% | +0.9 |
| 70–75% | 40 | 72.3% | 75.0% | 59.8–85.8% | +2.7 |
| 75–80% | 44 | 77.6% | 70.5% | 55.8–81.8% | −7.1 |
| 80–85% | 22 | 82.6% | 77.3% | 56.6–89.9% | −5.3 |
| 85–90% | 5 | 86.9% | 100.0% | 56.6–100.0% | +13.1 |
| 90–95% | 1 | 90.8% | 100.0% | 20.7–100.0% | +9.2 |
| All buckets | 515 | 59.7% | 65.4% | 61.2–69.4% | +5.7 |
The sportsbook series bins the same picks by the sportsbook’s own probability for the side we picked. Where that is under 50% the sportsbook had our pick as the underdog; those rows are the “under 50” bucket rather than being dropped.
The live season: each number was published before the card and graded after it. 421 graded picks from 24 January 2026 to 10 October 2026, of which 85 were published as toss-ups and are counted like any other pick.
| Bucket | n | Mean predicted | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| 50–55% | 84 | 52.5% | 50.0% | 39.5–60.5% | −2.5 |
| one voice | 78 | 52.4% | 50.0% | 39.2–60.8% | −2.4 |
| four-voice stack | 6 | 52.8% | 50.0% | 18.8–81.2% | −2.8 |
| debut bouts (market's pick) | 0 | — | — | — | — |
| 55–60% | 74 | 57.5% | 62.2% | 50.8–72.4% | +4.6 |
| one voice | 70 | 57.5% | 64.3% | 52.6–74.5% | +6.8 |
| four-voice stack | 4 | 58.1% | 25.0% | 4.6–69.9% | −33.1 |
| debut bouts (market's pick) | 0 | — | — | — | — |
| 60–65% | 63 | 62.5% | 73.0% | 61.0–82.4% | +10.5 |
| one voice | 54 | 62.5% | 72.2% | 59.1–82.4% | +9.7 |
| four-voice stack | 8 | 62.3% | 75.0% | 40.9–92.9% | +12.7 |
| debut bouts (market's pick) | 1 | 64.6% | 100.0% | 20.7–100.0% | +35.4 |
| 65–70% | 56 | 67.4% | 80.4% | 68.2–88.7% | +12.9 |
| one voice | 48 | 67.4% | 85.4% | 72.8–92.8% | +18.0 |
| four-voice stack | 8 | 67.4% | 50.0% | 21.5–78.5% | −17.4 |
| debut bouts (market's pick) | 0 | — | — | — | — |
| 70–75% | 62 | 72.6% | 82.3% | 71.0–89.8% | +9.7 |
| one voice | 49 | 72.6% | 85.7% | 73.3–92.9% | +13.1 |
| four-voice stack | 13 | 72.4% | 69.2% | 42.4–87.3% | −3.2 |
| debut bouts (market's pick) | 0 | — | — | — | — |
| 75–80% | 40 | 77.3% | 67.5% | 52.0–79.9% | −9.8 |
| one voice | 33 | 77.3% | 72.7% | 55.8–84.9% | −4.6 |
| four-voice stack | 7 | 77.4% | 42.9% | 15.8–75.0% | −34.5 |
| debut bouts (market's pick) | 0 | — | — | — | — |
| 80–85% | 34 | 82.2% | 85.3% | 69.9–93.6% | +3.1 |
| one voice | 18 | 82.2% | 83.3% | 60.8–94.2% | +1.1 |
| four-voice stack | 14 | 82.3% | 85.7% | 60.1–96.0% | +3.5 |
| debut bouts (market's pick) | 2 | 82.1% | 100.0% | 34.2–100.0% | +17.9 |
| 85–90% | 8 | 87.4% | 75.0% | 40.9–92.9% | −12.4 |
| one voice | 1 | 86.7% | 100.0% | 20.7–100.0% | +13.3 |
| four-voice stack | 6 | 87.3% | 66.7% | 30.0–90.3% | −20.6 |
| debut bouts (market's pick) | 1 | 89.1% | 100.0% | 20.7–100.0% | +10.9 |
| All buckets | 421 | 65.2% | 69.4% | 64.8–73.6% | +4.1 |
The served model changed on 5 September 2026; the record page discloses the same change. Up to and including the 29 August 2026 card every published probability came from a single voice; from the 5 September 2026 card it comes from all four together on every bout except the debut bouts below. The two are graded as two series, and each bucket in the table above splits the same way, because a single curve across both would describe a model this site no longer serves. The later series is 66 picks, so its intervals are wide and its bucket rows are indicative at best; it will firm up card by card.
From the 3 October 2026 card, a bout involving a fighter making a UFC debut is published with the market's own probability instead of a model's, and with no tier. Those 4 picks are counted in every total on this page, the 2026 panel and the all-years view included, and graded as a line of their own: they are in neither model series above.
Which of the two served a given pick is derived from its date: the graded archive carries no per-pick record of it, so the boundary used here is the day the serving path changed rather than a stamp on the row. 351 of the 421 2026 picks are assigned that way.
| Sportsbook bucket | n | Mean implied | Actual | 95% CI (Wilson) | Gap (pp) |
|---|---|---|---|---|---|
| under 50% | 32 | 44.4% | 40.6% | 25.5–57.7% | −3.8 |
| 50–55% | 80 | 51.0% | 61.3% | 50.3–71.2% | +10.2 |
| 55–60% | 41 | 57.9% | 65.9% | 50.5–78.4% | +7.9 |
| 60–65% | 62 | 62.9% | 62.9% | 50.5–73.8% | 0.0 |
| 65–70% | 46 | 67.6% | 65.2% | 50.8–77.3% | −2.4 |
| 70–75% | 47 | 72.5% | 80.9% | 67.5–89.6% | +8.4 |
| 75–80% | 49 | 77.3% | 79.6% | 66.4–88.5% | +2.3 |
| 80–85% | 28 | 82.3% | 89.3% | 72.8–96.3% | +7.0 |
| 85–90% | 22 | 87.4% | 90.9% | 72.2–97.5% | +3.6 |
| 90–95% | 13 | 92.4% | 84.6% | 57.8–95.7% | −7.8 |
| 95–100% | 1 | 95.4% | 100.0% | 20.7–100.0% | +4.6 |
| All buckets | 421 | 65.6% | 69.4% | 64.8–73.6% | +3.8 |
The sportsbook series bins the same picks by the sportsbook’s own probability for the side we picked. Where that is under 50% the sportsbook had our pick as the underdog; those rows are the “under 50” bucket rather than being dropped.
Until 16 September 2026 the market price we served could come from a prediction exchange as well as a sportsbook, so some 2026 rows in this series carry an exchange price; where our market number came from, before 2026-09-16 sets out how many and what we cannot establish.
The rows. Every graded pick on the record page’s loader, and only those: wins, losses and the toss-ups together, 1,953 in all. A toss-up is a graded pick with a published number and is counted like any other; leaving out the rows we were least sure of would flatter exactly the thing this page measures. The only rows that are absent are the ones the record page withholds because their source export is corrupt (none at this build).
The buckets. The 50% to 100% range is cut into ten equal-width probability buckets of five points each. A bucket with no picks in it is not drawn, so a chart can show fewer than ten dots: in this build the FightIQ series fills 8 of them on each chart, and the highest bucket drawn tops out at 90%. Each dot is one bucket, sized by how many picks it holds, at the bucket’s mean published probability across and its actual hit rate up; the whisker is a Wilson 95% interval on that hit rate. A bucket of seven picks that all won is not “100%”, it is “somewhere above 65%”, and the whisker says so.
Brier is the mean squared error of the published probability against the 0/1 result: lower is better, and publishing 50% on everything would score 0.2500. ECE (expected calibration error) is the n-weighted mean of the bucket gaps: 0 means every bucket sat on the diagonal. Both are recomputed from the archive at every build; nothing here is typed in.
Back-tested against live. 2023 to 2025 were scored after the fact against a placeholder freeze, which is a weaker claim than a live pick and is labelled that way on each panel. 2026 is the live season, published before each card. The methodology page sets out how the number is made and where it is weak.