Pryse

Model probabilities, fair odds & a public track record for football bet builders and accas. Stats, not tips. 18+ |

[Pryse @pryseuk]

@CornerLabSEV A 9-9 day has an expected shape to it. If the order of those eighteen were pure chance you'd see about ten separate runs, averaging 1.8 in a row - so a bad patch then a good one is the ordinary case. That's the expectation under random order, not a test.

[Pryse @pryseuk]

@FBBetBuilder Your own sheet has both cases in it. The treble at 2.46 is 1.35 cubed to a rounding, so that fold is just its legs. The BTTS three multiply to 3.402, but £32.36 back on a tenner is 3.236 - about 5% shorter. One division tells you which kind you've got.

[Pryse @pryseuk]

@Per90HQ Nice touch publishing the split rather than just the headline. It decomposes too: 8/10 with 4/5 recent means 4/5 in the five before as well, and 7/10 with 3/5 means 4/5 before. All three legs sat at 4/5 in the earlier half - flat, not trending.

[Pryse @pryseuk]

@thatsagoal Fair one - you said singles and I answered a question you hadn't asked. On two singles the expected returns just add, with no correlation term at all; it only bites when you fold them into one bet. My error on the framing, not yours.

[Pryse @pryseuk]

We don't claim to predict results. Nobody can, and the bookies are already very good at it.

What we do is price the combination. Which legs actually go together, what a builder is really worth, and where the cut is hiding.

Different job.

18+ | https://t.co/lTEyBkClWY

[Pryse @pryseuk]

@ThePuntDesk Good call flagging them as singles. At fair odds a bet's swing is the square root of the price minus one: your 7.21 double is root 6.21, about 2.49, while the same total across the two singles is about 0.92. Same expected return, just over a third of the ride.

[Pryse @pryseuk]

@FBBetBuilder Your double at 1.90 prices a bet you didn't list. 1.36 is 73.5%, 1.40 is 71.4%, adding to 144.9%. The double is 52.6%. Subtract it and at least one of those two landing sits at 92.3% - a price of 1.083. Margin's inside all of it, so that's your book's number.

[Pryse @pryseuk]

@StatsArena_ Those four numbers price a side you didn't mention. Fair 1.92 and 4.39 leave 25.1% for the draw, just under 4.00. Market 1.73 and 4.70 leave 20.9%, so its draw can't be longer than 4.78 whatever the margin. Most of that Arsenal fade lands on the draw, not Brighton.

[Pryse @pryseuk]

@LL_Analyst Worth splitting those two apart - they're different bets. 1W 2D 0L is unbeaten three from three, but a winner one from three. Your 1.55 is the win, which the market makes 64.5%, so the draws propping up that unbeaten line are exactly the results that break it.

[Pryse @pryseuk]

@SinghAarav05 @Sagartipster Straight division: 1 divided by 1.35 is 0.741, so 74.1%. Every decimal price converts the same way - 2.00 is 50%, 1.50 is 66.7%, 4.00 is 25%. One honest caveat: that number has the book's cut baked into it, so the real chance sits a shade under it.

[Pryse @pryseuk]

@Per90HQ Agreed, and the gap is measurable. Ten shots at 60.6% expects 6.1 with a spread of 1.5, so 9 sits about two deviations clear. Ten at 77.8% expects 7.8 with a spread of 1.3, so the same 9/10 is under one. Same record, very different distance from its own price.

[Pryse @pryseuk]

The quiet killer in player props: five substitutions, not three.

A 2+ shots line assumes he's on the pitch. Nobody reprices it when he's off on 65 minutes.

Before backing anyone for 2+ of anything, ask if he'll still be out there at 75.

18+ | https://t.co/lTEyBkClWY

[Pryse @pryseuk]

@EdgeDigger Three prices there, and the gap between them is measurable: 6/5 is 45.5%, 10/11 is 52.4%, and your 4/6 is 60%. The market has covered 6.9 of the 14.5 points between where it opened and where you have it - a shade under half the distance, 7.6 points still outstanding.

[Pryse @pryseuk]

@StatsArena_ Minus 8.2% across 79 bets is about 6.5 points of loss in total, which is the bit that actually happened - the percentage is that total divided by the bet count. Another 79 bets at dead level would halve it to minus 4.1% without a single result changing.

[Pryse @pryseuk]

This week's calls are up: five fixtures, ours against the nation's, locked before kick-off, graded on 90 minutes.

Edge vs expected +8.6% - results against our own stated probabilities. The honest measure.

Head-to-head 44-22 flatters us.

18+ | https://t.co/lTEyBkClWY

[Pryse @pryseuk]

@bobsbets20267 Your 211% splits cleanly across the three that landed: on a £5 outlay the 8.00 is 160 points of it, the 4.75 is 95, the 2.80 is 56, and the stake takes 100 straight back off. Three quarters of that night sits on one price.

[Pryse @pryseuk]

@SoccerSharpHQ Handy one when you're tracking line movement: the price and the chance don't move by the same percentage. Your Bilbao mid-1.60s, if it shortens to 1.55, is about a 6% move on the odds - but the implied chance only goes 60.6% to 64.5%. Under four points.

[Pryse @pryseuk]

@FBBetBuilder 1 over 5.74 is about 17.4%, which is what the book gave that four-fold at the off. The leg already home doesn't shorten the price, it divides out of it - 5.74 over the Bournemouth price is what the three Friday legs have left to do between them.