UFC Method + Round Combo Pricing: Breaking Down the Maths
Available in US
The market that punishes anyone who skips the maths
I used to eyeball method-plus-round combo prices. “KO in round two at 9.00? That feels about right.” Then I spent a weekend in 2018 reverse-engineering closing lines on a full card, and the number of combos that were priced 15% to 20% worse than their implied fair value genuinely surprised me. Not the bad ones — those were obvious. It was the mid-tier combos that looked reasonable on the surface and quietly ate expected value every time I backed them.
Combo markets are the place where bookmakers recover the margin they do not get to charge on clean moneyline bets. UK moneyline MMA sits at margins that are competitive with any sport. Combo markets can run overrounds of 8% to 12% or higher, and that cushion is where the book makes its money on punters who like building specific, picturesque tickets.
This piece is the maths I wish I had sat down with a decade ago. By the end, you should be able to reconstruct the fair price of any method-plus-round combo from the single markets, spot combos priced well beyond fair vig, and walk away from the ones that are just dressed-up sucker bets. If you want the broader context on method and round markets, the method and round betting guide has the full map.
The two-dice model for thinking about combos
A method-plus-round combo is two random variables glued together. One die decides how the fight ends — KO, submission, decision. The other die decides when it ends — round one, round two, round three, and so on. The combo ticket pays only if both dice land on the specific combination you chose.
If the two dice were truly independent, pricing would be trivial. You would multiply the probability of the finish type by the probability of the round, convert to odds, and add whatever overround the book wants to bake in. Fights do not work that way. The two dice are correlated, and the correlation is the entire interesting part of the market.
Consider KO in round one. The probability of a KO across all UFC fights sits at roughly 33%. The probability of a round-one finish of any kind sits at roughly 30% across the full dataset. If the two were independent, KO-in-round-one would be 0.33 × 0.30 = 9.9%, or fair odds of about 10.1. The actual rate of KO-in-round-one is higher than that because KOs cluster in the early rounds when fighters are fresh and willing to engage. The two-dice independence assumption under-prices KO-in-round-one relative to reality.
Submission, by contrast, distributes more evenly across rounds with a mild lean towards rounds two and three when grappling exchanges develop. Submission-in-round-one is therefore slightly over-priced by the independence model at most books — a detail worth knowing. The overall UFC submission rate sits around 20% and skews to the head and to choking submissions, which matters for how combos price at the weight classes with the most wrestling.
Conditional probability without the jargon
The sharper way to think about combo pricing is conditional. Given that a fight ends in round one, what is the probability that it ends by KO versus by submission versus (obviously, zero) by decision? Given that a fight ends by KO, what is the probability that it ends in round one versus round two versus round three?
Those are two separate breakdowns that give you the same combo price. Either you take P(round 1) and multiply by P(KO | round 1), or you take P(KO) and multiply by P(round 1 | KO). If the two computations disagree, the book’s combo line is not internally consistent and one of their single markets is mispriced. You can occasionally spot this on individual cards, and when you do, it is usually a signal about which single market the book cares less about pricing tightly.
The rough conditional rates I carry in my head are these. Given a KO finish, about 45% to 50% land in round one, 25% to 30% in round two, and the remainder in later rounds. Given a submission finish, round one accounts for about 30% to 35%, round two for another 30% to 35%, and later rounds fill in the rest. Given a decision, every round contributes equally by definition because decisions require reaching the final bell.
Apply those conditional numbers to your single market prices and you can rebuild the fair combo line. Any published combo price that sits more than a few percent above fair is vig. Any that sits at or below is a genuine opportunity and worth attention.
The overround check that takes sixty seconds
The quickest sanity check on combo pricing is a total-book test. Add up the implied probabilities of every method-plus-round combo on offer for a single fight. For a three-round bout, that is typically nine combos: KO/TKO in rounds 1-2-3, submission in rounds 1-2-3, plus decision and sometimes a DQ or “other” bucket. The sum of the implied probabilities gives you the total overround on the combo market for that fight.
A fair book with no vig would sum to 100%. Real books sum to 108% to 115% on combo markets, with 110% sitting as roughly the median across the UK-licensed sportsbooks I track. If you find a fight where the combo market sums to 105% or lower, somebody has published a loose line and there is probably specific value hiding in one of the combos. If it sums to 118% or higher, every combo on offer is carrying a cost and backing any of them is an exercise in buying pain.
The market is priced by computers at all the major UK books, which means loose lines get tightened fast. Most of the genuine value in the combo market lives in the first few hours after lines open, particularly for smaller undercard bouts where the algorithmic pricing is less attentive. If you have access to early prices and the discipline to place real money on them, that window is where disciplined combo punters genuinely outperform.
A step-by-step example
Let me walk through an example with invented-but-plausible numbers to show the method working. The fight is a welterweight three-round bout. Fighter A is a pressure-forward knockout striker. Fighter B is a durable technician.
Single market prices at the book. Fighter A by KO/TKO sits at 2.50 (implied 40%). Fighter A by submission at 15.00 (7%). Fighter A by decision at 3.50 (29%). Fighter B by KO/TKO at 11.00 (9%). Fighter B by submission at 17.00 (6%). Fighter B by decision at 6.00 (17%). Sum of all methods is 108%, which is a normal 8% overround on method. Goes-the-distance yes is priced at 2.15 (47%) and no at 1.80 (55%), summing to 102% — a tight market.
Rebuilding fighter A by KO in round one. We start with 40% KO probability. Given a KO finish, roughly 47% should land in round one given the early-round KO clustering. The unconditional probability of fighter A’s R1 KO is therefore about 0.40 × 0.47 = 18.8%, or fair odds of 5.32. If the book’s published price on “fighter A to win by KO in round 1” sits at 6.00, we have found roughly 13% above fair vig — a strong combo by combo-market standards. If the book has it at 4.50, we are paying vig and the combo is a pass.
Do that same calculation for each of the eighteen or so combos on offer for the fight. Two or three will usually sit at fair-to-generous prices. The rest will sit at vig-loaded prices that are fine on taste but unprofitable on math. Stick to the two or three and skip the rest, and combo betting becomes a genuine positive-expectation exercise instead of a slow bleed.
The habit that makes the maths pay
I do not calculate combo fair value in my head. I have a spreadsheet with the conditional probability table I described above, and I punch in the single-market prices for any combo I am considering. The whole exercise takes under a minute per fight and it is the only reason I still come out ahead on combo markets.
The bettors who lose steadily on combos are the ones who find a combo that “feels right” and place it without checking. The bettors who do fine are the ones who start from the single markets, rebuild the fair combo, compare it to the published combo, and only stake when the published price is at or above fair. If that sounds tedious, it is, and that is exactly why most of the market does not do it and exactly why the edge exists for anyone who will.
Is a method-plus-round combo independent or conditional?
Conditional. The probability of a specific finish type interacts with the round distribution in ways that are not independent. KOs cluster early. Submissions distribute more evenly with a mild middle-round peak. Decisions are by definition confined to the final round. Anyone pricing combos as a flat multiplication of method-price by round-price will consistently misprice the early-round KO and the late-round submission in particular.
How much vig sits on combo markets?
Typically 108% to 115% total implied probability across all combos for a single fight at UK-licensed books, with 110% as the rough median. Compare that to moneyline vig of around 102% to 105% and method-of-victory vig of around 107% to 110%. Combos sit at the higher end of the vig scale and reward more careful fair-value reconstruction before staking.
Can I always reconstruct the combo from singles?
Not perfectly, but closely enough to be useful. The singles markets give you the method distribution and the round distribution independently. Conditional adjustments — how KO finishes cluster early, how submissions spread more evenly — are the educated-guess layer on top. The reconstructed fair price is usually within 5% of the true fair price, which is tight enough to identify the combos that are priced well above fair vig and therefore actually worth backing.
Written by the editors at Round Betting ufc.
