The second half of March is in the books. 6,577 matches, $37.3 billion in Salty Bucks, and one of the most chaotic stretches we’ve recorded. Let’s check it out!
| Stat | March 16–31, 2026 |
|---|---|
| Total matches | 6,577 |
| Unique fighters seen | 6,036 |
| Total Salty Bucks wagered | $37.3 billion |
| Average pot | $5.7 million |
| Red wins | 3,241 (49.3%) |
| Blue wins | 3,336 (50.7%) |
Combined with the first half (5,086 matches), March 2026 totaled 11,663 matches — a healthy month, right in line with the ~12,500/month average.
Red vs Blue continues its eternal non-rivalry: 49.3% vs 50.7%. Blue edges as always. At this point we’ve said it enough — neither side has a meaningful advantage, and we’ll probably keep saying it because people keep asking.
| Tier | Matches | Avg Pot |
|---|---|---|
| A | 2,823 (42.9%) | $5.2M |
| S | 2,098 (31.9%) | $6.5M |
| B | 1,346 (20.5%) | $5.0M |
| X | 170 (2.6%) | $6.6M |
| P | 140 (2.1%) | $8.6M |
The usual distribution. A-tier continues to be the workhorse at 43% of all matches. P-tier is interesting — only 140 matches but the highest average pot at $8.6M. The crowd bets big on P-tier fights, probably because the outcomes feel more uncertain (and they are — P-tier has the lowest crowd accuracy historically).
These fighters went the entire second half of March without a single loss:
| Fighter | Record | Tier |
|---|---|---|
| Chie satonaka | 9-0 | A |
| Rockman_exe | 8-0 | S |
| Demon king dhaos | 7-0 | S |
| Evil zangief hd | 7-0 | S |
| Goodman | 6-0 | A |
| Houou jmf | 6-0 | S |
| Justice gg1 ex | 6-0 | S |
| Kaban | 6-0 | B |
Chie Satonaka leads the pack at 9-0 in A-tier — the Persona 4 rep is on an absolute tear. Rockman_exe (Mega Man Battle Network) follows at 8-0 in S-tier, which is impressive given the level of competition. Playing against Zangief would be brutal — those grabs and throws would drive me insane.
Worth noting: four of the top eight are S-tier. Going undefeated in S-tier — even for a short stretch — means you’re beating the best fighters in SaltyBet. These aren’t soft schedules.
Not everyone had a good March:
| Fighter | Record | Tier |
|---|---|---|
| Adol christin | 0-9 | S |
| Akebono taro | 0-7 | B |
| Nono | 0-6 | S |
| Reimu-q | 0-6 | S |
| Ermac | 0-5 | A |
| Goku ssjgss | 0-5 | S |
Ermac! No! My sweet Ermac! Adol Christin — the hero of Ys, a legendary RPG franchise — lost 9 straight in S-tier. Goku SSJGSS (Super Saiyan God Super Saiyan, lol) also went 0-5. As we noted in our piece on why some MUGEN characters always win, the name on the select screen doesn’t always match the performance in the arena.
| Winner | Defeated | Odds Against | Tier |
|---|---|---|---|
| C-xion | Saber_alter | 24.3:1 | S |
| Supreme goku | Kabbu-ki! | 22.1:1 | X |
| Jagi bike | Nayuki the runner | 18.9:1 | A |
| Hirasaka hina | Axel-hell | 16.5:1 | S |
| Leon c | Q-bee ex | 14.0:1 | A |
C-xion pulled off the biggest upset of the half — beating Saber_alter at 24:1 odds in S-tier! Twenty four to one! When the crowd bets 24 times more money on your opponent and you still win, that’s the kind of moment that makes SaltyBet worth watching.
Supreme Goku’s 22:1 upset in X-tier is notable because X-tier upsets are genuinely rare. The crowd has the most information about X-tier fighters, so when they’re that wrong, something unusual happened.
How did SaltyTrack’s ML model perform in the second half of March?
| Confidence | Correct | Total | Accuracy |
|---|---|---|---|
| Lock | 8 | 8 | 100.0% |
| High | 1,297 | 1,526 | 85.0% |
| Medium | 1,902 | 2,822 | 67.4% |
| Low | 1,310 | 2,221 | 59.0% |
| Overall | 4,517 | 6,577 | 68.7% |
Lock predictions continue to be perfect — 8 for 8, cash me out. High confidence holds strong at 85.0%. Overall accuracy came in at 68.7%, which is below our typical ~72% — working on it boss!
Here’s the interesting part: the crowd came in at 68.5% for the same period — essentially identical to the model’s 68.7%. Normally we edge the crowd by 3-4 points. When both the model and the crowd converge like this, it means the fights themselves were harder to predict. The second half of March had a higher-than-average share of genuinely unpredictable matchups, which the Chaos Score data confirms — won’t stop me from continuing to optimize the model, though, gotta give you all the competitive edge, feel me?!
Speaking of chaos — the second half of March had some wild days:
| Day | Chaos Score | Upsets | High-Conf Wrong |
|---|---|---|---|
| March 16 | 433 | 140/426 (32.9%) | 33 |
| March 19 | 393 | 143/418 (34.2%) | 29 |
| March 17 | 351 | 123/386 (31.9%) | 25 |
| Calmest: March 23 | 168 | — | — |
March 16 scored 433 — that’s the 6th highest Chaos Score in our entire all-time dataset! 33 high-confidence predictions failed in a single day. The stretch from March 16–19 was one of the most chaotic 4-day runs we’ve ever recorded, with three consecutive days above 350. Madness!
March 23 was the calm after the storm at 168 — a quiet Sunday.
Combining both halves:
| Stat | Full Month |
|---|---|
| Total matches | 11,663 |
| Total wagered | ~$70 billion |
| Crowd accuracy | ~68% |
| SaltyTrack accuracy | ~70% |
| Most chaotic day | March 16 (score: 433) |
| Undefeated highlight | Chie Satonaka (9-0, A-tier) |
| Biggest upset | C-xion over Saber_alter (24:1, S-tier) |
April’s meta report drops in two weeks. In the meantime:
See these stats live during matches. The SaltyTrack Chrome Extension shows you win rates, head-to-head records, and AI predictions right on saltybet.com. Free, no account needed.
SaltyBet uses virtual currency only. No real money is wagered or exchanged. SaltyTrack is not affiliated with SaltyBet.