Keith Berish
- Fighting out of
- Poughkeepsie
- Born
- United States
- Camp
- Black & Blue MMA
- Weight class
- Middleweight
- Height
- 6'0" · 183 cm
- Reach
- 74" · 187 cm
- Stance
- Orthodox
- Last fight
- Dec 9, 2016
UFC fight history
Outside the UFC
| Res | Opponent | Outcome | Finish | Rd | Time | Promotion | Date |
|---|---|---|---|---|---|---|---|
| W | Ariel Sepulveda6-7 | Submission | Rear-Naked Choke | R1 | 4:30 | ROC 47 | |
| W | Patrick Walsh10-3 | Submission | Rear-Naked Choke | R1 | 0:18 | VCS | |
| W | Michael Andrillo1-3 | KO/TKO | Arm Injury | R1 | 1:10 | ROC 43 | |
| W | David Colabella1-2 | KO/TKO | Punches | R1 | 1:04 | ROC 36 | |
| W | Phil Dace9-5 | Decision | Unanimous | R3 | 3:00 | Reality Fighting |
Non-UFC bouts (regional/other promotions) and Dana White’s Contender Series — shown for context only, excluded from Schedule Score, striking/grappling rates, and every UFC-scoped metric.
Method of victory & defeat
KO history
Knockout power and chin durability (UFC; clean KOs only).
Never finished by strikes · Never submittedin the UFC
Judging
Scorecard data ingested from MMA Decisions
Career striking
Striking zones
Per-15-minute breakdown — where this fighter lands, and where they get hit.
Ranks are division percentiles — e.g. 90th = top 10% of the weight class. Higher is always better: on “absorbed” rows the rank is inverted, so a high rank means they get hit less.
Ranks are division percentiles — e.g. 90th = top 10% of the weight class. Higher is always better: on “absorbed” rows the rank is inverted, so a high rank means they get hit less.
Career totals — landed / attempted
| Zone | Per 15 | Landed | Attempted | Accuracy |
|---|---|---|---|---|
| Head | 39.0 | 39 | 136 | 29% |
| Body | 6.0 | 6 | 13 | 46% |
| Leg | 16.0 | 16 | 19 | 84% |
| Zone | Per 15 | Landed | Attempted | Accuracy |
|---|---|---|---|---|
| Head | 55.0 | 55 | 180 | 31% |
| Body | 15.0 | 15 | 21 | 71% |
| Leg | 12.0 | 12 | 13 | 92% |
UFCStats career rates, per 15 minutes. Percentiles vs the division; absorbed is inverted so higher is better.
Grappling
Per-15-minute breakdown of takedowns, submissions, and ground control.
Career totals — landed / attempted, control time
| Takedowns | 2 / 5 |
| Submission attempts | 0 |
| Control time | 0:52 |
| TDs defended | 0 / 0 |
| Control absorbed | 2:59 |
Rates above are per 15 minutes of UFC fight time (15min total). Value colors reflect the per-15 rate’s percentile against the current weight class baseline; raw totals without a division reference stay neutral.
Pace & fade
Round-by-round work rate, and whether they fade.
Berish hasn't reached the third round in enough UFC bouts to read a reliable pace pattern — most of their fights end early. The round-by-round pace below is shown for what rounds there are.
Faded points are drawn from a single bout that reached that round, so they’re less reliable than the solid points averaged over many fights.
Rounds backed by fewer than 2 bouts are drawn smaller and dimmed with an n= count — a deep round seen in one or two fights is a thin sample, not a settled rate. Per-round bout counts are in the breakdown below.
Full per-round breakdown
What drives the pace · per round
The composite output line above blends these three inputs — splitting them out shows whether the work rate is built on striking volume, takedown pressure, or top control, and which input fades.
| Round | Output / min | Sig landed / min | Sig absorbed / min | TD att / min | Control | Bouts |
|---|---|---|---|---|---|---|
| R1 | 6.8 | 6.2 | 7.0 | 0.20 | 0% | 1 |
| R2 | 4.8 | 4.2 | 5.4 | 0.20 | 0% | 1 |
| R3 | 4.3 | 1.8 | 4.0 | 0.60 | 17% | 1 |
Output / minis the headline work rate — significant strikes landed, plus 3 per takedown attempt, plus one per 15 seconds of control — so a takedown-and-grind round isn’t scored as idle the way bare striking volume would. The remaining columns are its components. Per-round rates use UFCStats per-round data (available for ~95% of bouts since 2008). Each round’s rates are averaged only over bouts that reached that round, so the bout count shrinks as rounds deepen. The final round of a finished fight is pro-rated by its actual length.
Schedule Score
Opponent quality, ranked across the UFC. Recent fights weighted heavier.
- Building3+ UFC fights, ≥.40 win rate
- UnprovenFewer than 3 fights, or sub-.40 win rate
| Tier | Bouts | Record | Win rate |
|---|---|---|---|
| Building | 1 | 0-1 | 0% |
| Unproven | 1 | 0-0, 1 NC | — |
A descriptive read of the non-UFC slate from opponents’ full career pro records — not a UFC-grade Schedule Score.
All bouts — opponent quality bout-by-bout
| Res | Opponent | Tier | Score | Date |
|---|---|---|---|---|
| L | Ryan Janes | Building | 15 | Dec 9, 2016 |
| NC | Robert Drysdale | Unproven | 8 | Jul 6, 2014 |
Per-bout quality scoring: champion = 100; same-division ranks 1–5 = 90 → 78 linear; ranks 6–15 = 75 → 60 linear; cross-division champion = 78; cross-division ranked = 45; title-fight challenger (no rank signal) = 50, bumped to 65 when opp had an elite career UFC record at the time. Unranked opponents are scored on UFC record-at-time (long-tenured vet 60, contender-level 45, established 30, building 18, unproven 8). P4P top-10 at time adds +5; top-3 adds +8 (capped at 100). The Schedule Score combines schedule strength (65%) and quality wins (35%), then is held toward a rookie baseline for fighters with limited UFC experience so a thin résumé reads honestly low.
Career cohorts
1 resolved UFC boutsPerformance broken down by situation — camp length, response to the previous result, title-fight presence, and rematches. Computed across every completed UFC bout.
Activity & layoff
Last fight, activity, and divisional context.
DraftKings fantasy scoring
DFSAs underdog 0-1See Keith Berish's DK score in every fight
The full fantasy-points log with closing lines — sortable, filterable, with a live summary — for all 1 UFC bouts.
Unlock with Pro →Sources: UFCStats, Sherdog, FightOdds.io, Kalshi.