Striking vs Grappling: How Mixed Martial Arts Matchups Resolve

Where each one is fragile

The striker's fragility is compounding. Every failed takedown defence costs energy, and energy is what makes the next sprawl work. A striker who defends four takedowns and fails the fifth is often in far worse shape than the raw count suggests, because the fifth one lands against a tired body.

The grappler's fragility is front-loaded. Entries happen at striking range, which means a grappler has to travel through the striker's best real estate to reach their own. A shot taken from too far out is a shot taken with the head low and the chin exposed.

Neither fragility is a flaw to be coached away. They are the costs each approach pays for the control it seeks.

What the judges do with it

The scoring layer matters here, because a lot of fights in this matchup go to a decision.

Judging guidance treats effective striking and effective grappling together as the primary criterion, with legal strikes that inflict damage counted alongside successful takedowns, reversals and submission attempts. Judges are also asked to take each round on its own merits rather than weigh the cumulative impact of strikes across a fight.

The practical reading is that grappling scores when it is offensive. A takedown followed by sustained work scores; a takedown followed by stalling in a neutral position is a much weaker case, and it is the single most common source of arguments about a scorecard. How MMA judging criteria work goes through the hierarchy in detail.

Deciding where the fight happens first

The biggest realism fix we ever made to 40-0 MMA & Boxing Fight Draft was not an attribute change. It was reordering the resolution.

The original model resolved an exchange by comparing relevant attributes and producing an outcome. It weighed striking, wrestling, takedowns, submissions and ground control, all in one pass. It looked reasonable and it was badly wrong in one specific way: it systematically overrated strikers. A fighter with excellent striking numbers and poor takedown defence would beat a wrestler more often than they should, because their striking numbers were contributing to every exchange regardless of whether the exchange was one they should have been allowed to have.

What shipped resolves in two stages. First, the model decides the phase — does this exchange happen standing, in the clinch, or on the ground — using takedown ability against takedown defence, with fatigue applied. Only then does it resolve the exchange, using the attributes that are actually relevant to that phase. A striker's striking is worth nothing in an exchange that is happening from bottom position.

The change in behaviour was immediate and, for a while, unpopular with testers who had built stables of high-rated strikers. It is also the closest the model has come to reproducing something the sport genuinely does, which is why it stayed.

A limitation worth naming. The phase decision is made per exchange, not continuously, so the simulator represents a fight as a sequence of discrete situations rather than as a flowing contest. Scrambles — the transitional moments where neither fighter has position and the fight is genuinely in between phases — are the thing this model cannot express. They are resolved into whichever phase came next, which loses a real and important part of the sport.

How much further that style modelling goes, and where we decided to stop, is in modelling style matchups in a fight simulator. The weight-class constraints that bound every matchup are in MMA weight classes.

The game's fighter pools are built from real fighters' careers, with ratings and records that are the game's own estimates; it is not affiliated with, endorsed by or licensed by any promotion, sanctioning body or athlete, and no result describes or predicts a real bout. The studio's other combat and sports titles are under sports GM.