If you have never bet in Poland, this one will look strange. Every legal bet placed with a licensed Polish operator carries a 12% tax on the stake — not on winnings, but on what you put in. It is not a fee you pay separately: the bookmaker folds it into the price, so what you see is already “after tax”. The effect, though, is real and countable.
A simple example
Take odds of 2.00 on an event with a true probability of 50%. With no margin and no tax that would be a “fair” bet — over the long run you break even.
You stake 100 PLN. But the 12% tax means only 88 PLN actually goes into play (12 PLN goes to the treasury). Your effective price is therefore not 2.00 but:
effective odds = nominal odds × 0.88 = 1.76
Put differently: to break even at a displayed price of 2.00, the event has to occur not 50% of the time, but roughly 57% of the time. That gap is your fixed cost of participation — before we even add the bookmaker’s margin.
It applies at every price
The ”× 0.88” rule is constant, but its sting is not spread evenly across the board. A few examples of effective odds (nominal price → what you actually get):
1.50 → 1.32 · 2.00 → 1.76 · 3.00 → 2.64 · 5.00 → 4.40 · 10.0 → 8.80
In percentage terms the cut is always the same (12%), but at short prices it pushes a bet below the break-even threshold faster — because the margin for error there is narrow to begin with. On “favourites” in the 1.20–1.50 range, the tax alone can turn a slightly positive situation into a clearly negative one.
Tax is not the only cost
The tax is merely the first layer. The second is the bookmaker’s margin (overround) — the surcharge built into the odds that makes the probabilities of all outcomes add up to more than 100%. The two costs compound: first the operator takes its margin out of the “clean” price, and then 12% tax comes off your stake.
So the real distance between you and break-even is greater than the tax alone suggests. A bettor in Poland is not playing at a “fair” price, nor even at the closing line — they are playing at that price minus the operator’s margin and minus the turnover tax.
Why this matters when measuring accuracy
When we compare the model’s probabilities with prices, we always do it against the closing line (the final price before kick-off), because that is the sharpest, best-informed price the market produces. It is a purely analytical benchmark. But nobody bets “at the closing line, free of costs”.
This is one of the reasons why naively betting the model’s deviations produces negative ROI. The model may indicate that a price is 3% too high — but the tax alone is 12% at the door. The discrepancy would have to be genuinely large and genuinely real to break through that barrier. Usually it is neither.
What to do about it
Nothing magic — this is arithmetic, not a loophole to exploit. Awareness of the cost is the first step towards measuring your own decisions honestly. That is why at LigaMetrics the bettor’s tax is an explicit parameter of every analysis rather than a hidden constant: when we show a model–market discrepancy, we show it next to the costs you would have to overcome anyway. Not to encourage betting — so that the numbers do not lie.
Educational material. Not advice on taking part in betting. 18+.