The Parlay Tax Index
Every prop you bet has a fee baked into the price, and it is not the same fee on every market. We measured it on 73,872 two-sided prop prices captured between 2026-05-27 and 2026-08-25. The cheapest market on the board charges 5.98%. The priciest charges 7.02%. Nobody itemizes this on your slip, so here it is.
| Market | Sport | Book keeps | Per $100 | Prices |
|---|---|---|---|---|
| Pitcher outs recorded | MLB | 7.02% | $7.02 | 2,415 |
| Assists | NBA/WNBA | 7.02% | $7.02 | 851 |
| Points + assists | NBA/WNBA | 7.01% | $7.01 | 1,009 |
| Points + rebounds | NBA/WNBA | 7.01% | $7.01 | 1,277 |
| Rebounds | NBA/WNBA | 7.01% | $7.01 | 1,198 |
| Three-pointers made | NBA/WNBA | 7.01% | $7.01 | 893 |
| Hits + runs + RBIs | MLB | 6.98% | $6.98 | 25,134 |
| Points + rebounds + assists | NBA/WNBA | 6.96% | $6.96 | 1,068 |
| Batter total bases | MLB | 6.92% | $6.92 | 11,179 |
| Batter hits | MLB | 6.84% | $6.84 | 24,719 |
| Points | NBA/WNBA | 6.53% | $6.53 | 1,476 |
| Pitcher strikeouts | MLB | 5.98% | $5.98 | 2,653 |
Median margin per market. Bars are scaled between the cheapest and priciest markets, not from zero — the whole range is about one percentage point, so a zero-based bar would render every row identical. Markets with fewer than 200 measured prices are excluded.
What the number means
Take any prop and look at both prices. Convert each to a percentage: a −110 is 52.4%, a +150 is 40%. Now add the two sides together. If the market were fair they would sum to exactly 100%, because one of them has to happen. They never do. They sum to something like 107%, and that extra 7% is the house’s cut — charged whether you win or lose, on every single bet, without ever appearing as a line item.
That is the whole calculation. It is not complicated, it is just never shown to you, and it is the reason a bettor who picks correctly slightly more than half the time still loses money.
The finding
Most markets sit close together, between roughly 6.8% and 7.0%. The interesting part is the outlier: pitcher strikeouts are meaningfully cheaper than everything else at 5.98%, measured across 2,653 prices.
1.04 percentage points sounds small. It is not. Betting the priciest market instead of the cheapest costs you about 17% more in fees on identical stakes, forever, regardless of whether your picks are any good. That is a decision you make before you have handicapped anything, and most people make it without knowing there was a decision.
The reason is liquidity. The more money and attention a market attracts, the tighter a book has to price it to stay competitive. Thin, novelty and combination markets carry more margin because fewer people are shopping them. That is also why combining legs into a parlay compounds the problem: you pay the margin on each leg, and the margins multiply rather than average.
How we measured it
Every price our pipeline captures is stored with a timestamp. For this index we used only markets where both sides were posted, since you cannot compute a margin from one side alone — that left 73,872 prices from 2026-05-27 to 2026-08-25. Both sides were converted to implied probability and summed, and the excess over 100% recorded as the margin.
We report the median, not the average, so that a handful of extreme longshot prices cannot drag a market’s number around. Markets with fewer than 200 measured prices are left out entirely rather than shown with a shaky number.
What this does not tell you
It does not compare sportsbooks. You would expect that to be the headline, and it is deliberately absent. Until 2026-08-25 our price history kept one book per prop, which means the rows we hold for different books describe different props rather than the same prop priced twice. Comparing them would be comparing bet mixes and calling it pricing. We started capturing every book’s price on 2026-08-25; the book-by-book table gets added when there is enough of it to be honest, and not before.
It also is not a prediction. A market being cheap does not make a bet in it good — it makes the toll smaller. What you do after that is still on you.
FAQ
What is the parlay tax?
The sportsbook’s built-in margin. Add both sides of any market together and they sum to more than 100%; the excess is what the house keeps. Across the prices we measured, the median runs between 5.98% and 7.02% depending on the market.
Which prop market has the lowest margin?
Pitcher strikeouts — 5.98% median across 2,653 prices. The priciest we measured was pitcher outs recorded at 7.02%.
Does a lower margin mean I will win?
No. It means less is taken off the top before the bet is graded. Nothing here predicts an outcome, and anyone promising you a winning bet is selling something. Lower fees are a real edge over a long enough run, and they are the only part of this you fully control.
Can I download the underlying prices?
No. The odds feed behind this is licensed for use inside an application, not for redistribution, so we publish the analysis and never the raw prices. The method is described above in full, which is the part worth having anyway.
Want to see what this costs on a slip you are actually looking at? The parlay payout calculator takes your stake and each leg’s price and shows what the book keeps out of that exact ticket. Or read why the margin compounds on every leg.
ET Parlays is a sports-analytics research tool, not a sportsbook. This is research and education, not betting advice. 21+, and only ever bet what you can afford to lose. See responsible gambling resources.