Implied Probability and Vig on MLB Outright Boards

What Implied Probability Tells You

I once bet a friend a pint that he couldn’t tell me what 9/2 meant in plain English without using the word “odds”. He couldn’t. Most punters can’t. They can read the price, work out the return, even spot a drift – but ask them to translate a number into a probability and the room goes quiet.

That’s the gap implied probability fills. Every odds line is a bookmaker’s bet on how likely the outcome is, expressed in the language of payout rather than percentage. Convert it back to percentage and you’re suddenly speaking the same language as the trader who set the price. That’s where outright betting starts being a contest of opinions rather than a guess at numbers.

For an MLB outright market – World Series, pennants, division winners, MVPs – the conversion matters more than for most sports, because the field is wide and the prices spread across a huge probability range. A 30-team World Series board has a favourite at maybe 15% implied and the basement teams at under 1%. You can’t compare those positions or hedge across them without working in percentages.

Converting Decimal to Implied

The simplest conversion in betting maths is decimal to implied probability: 1 divided by the decimal price. A 5.00 price is 1/5, which is 20%. A 7.50 price is 1/7.50, which is 13.3%. A 33.00 price is 1/33, which is 3.03%. Pull up any decimal-formatted board and you can read implied probability off the top of your head once you’ve done this for a week.

Fractional needs one extra step. The denominator divided by the sum of numerator and denominator gives implied probability. So 13/2 is 2 divided by (13 plus 2), or 2/15, which is 13.3%. American is two formulas depending on sign – for positives, 100 divided by (American plus 100); for negatives, the absolute value divided by (absolute value plus 100). +650 is 100/750 = 13.3%. -200 is 200/300 = 66.7%. The arithmetic is rote, and after a fortnight of working through your bet slips you’ll do it without noticing.

Here’s the part most punters miss. The implied probability you calculate from any single price is not the bookmaker’s actual estimate of probability. It’s their estimate plus a margin built in to guarantee profit. Bill Miller of the American Gaming Association called 2025 a strong year for the industry, with prediction markets shaping the defining regulatory fight, and the entire reason that fight matters is because that built-in margin – the vig, the juice, the overround – is the bookmaker’s revenue model. To see what the bookmaker actually thinks is going to happen, you have to strip the margin out.

Normalising the 30-Team Board to 100%

Sum the implied probabilities of every single team on a World Series outright board and you’ll get a number bigger than 100%. On a typical UK pennant market the figure lands somewhere between 110% and 115%. On the busier multi-way MLB markets at retail US sportsbooks, you can see the overround push past 130%. That excess is the vig – the bookmaker’s pre-event take-out. The hold on a standard -110/-110 two-way US line works out to 4.76% on the bookmaker’s side, with the punter’s edge sitting at 4.55% before any skill is applied. NFL futures field-wide hold runs as high as 12%. MLB sits between those numbers, depending on the depth of the field.

To get the bookmaker’s actual implied probability for any team, you normalise. Divide the team’s raw implied probability by the total of the board, then multiply by 100. So if the Dodgers sit at 13.3% raw and the full World Series board sums to 115%, their normalised probability is 13.3 divided by 1.15, which is 11.6%. That 11.6% is what the trader is really betting on – the 13.3% is what they want you to pay.

The gap between raw and normalised is where the punter’s cost lives. Across a six-month outright position, that 1.7 percentage points compounds into real money. If you’re consistently backing 30-team boards without normalising, you’re paying tuition fees you didn’t know you’d signed up for. The same logic applies in reverse on a 12-team postseason field – narrower field, narrower hold, but still meaningful.

I run a normalised column on every outright board I’m watching seriously. It takes ten minutes in a spreadsheet to set up and it shows you, at a glance, where the trader’s view diverges from the price you’re being shown. That divergence is the only thing worth betting on.

The Power Method, More Accurate

Simple normalisation has a flaw, and it’s worth understanding even if you don’t use the fix. Dividing every price by the same overround assumes the bookmaker spreads the margin evenly across the board. They don’t. They load more margin onto the longshots, where casual punters lump money sentimentally, and less onto the favourites, where sharp money keeps prices honest.

The power method corrects for this. Instead of dividing every implied probability by the same number, you raise each implied probability to a power between 0 and 1, choosing the power that makes the resulting probabilities sum to exactly 100%. The favourites barely move; the longshots get pushed down hard. For a typical MLB World Series board, the power method might give you a Dodgers normalised probability that’s almost identical to simple normalisation, but a Pirates probability that’s notably lower than simple normalisation suggests.

You don’t need to memorise the formula. Most outright spreadsheets sold to or built by serious punters bake the power method in. What you need to know is that simple normalisation overstates longshot value and understates favourite value, and that the power method is the standard correction. When you see a deep dive on whether a 50/1 outsider offers genuine value, the analyst has almost always run the power method already – the gap between simple-normalised and power-normalised probabilities for that team is usually the headline finding.

Spreadsheet Templates UK Punters Use

The template I use has nothing exotic in it. Eight columns. Team name. Price in decimal. Implied probability – that’s =1/B2. Sum row at the bottom for the implied column, which tells you the overround. Normalised simple – that’s implied divided by the sum row. Normalised power – that’s a slightly fiddlier formula using SOLVER, but copy-paste from any template online and you’re done. Then two columns for your own probability estimate and edge: edge is your probability minus normalised probability, and any positive number is a candidate bet.

I lock the price column to live updates if I’m watching a market closely – most exchanges and a couple of the bigger UK bookmakers have data feeds that work with Google Sheets if you know where to plug them in. For pre-season outright work, manual entry once a week is plenty. For deadline-week pennant repricing or post-injury news, you want it more responsive than that.

The single most useful column is none of the maths. It’s a notes column where I jot the date and the reason a price moved. Three months later, when the same line drifts again, that notes column tells me whether the previous move was justified by the outcome or whether the market overreacted. Pattern recognition across a season is impossible without it.

Reading the Board Like a Trader

Once implied probability and normalisation are part of how you read prices, the outright board stops looking like a list of bets and starts looking like a published opinion. Every line is the bookmaker telling you, in numerical form, what they think the world is going to do for the next six months. Every line is also the bookmaker telling you what they’re charging you for the privilege of disagreeing.

Your job from here is to disagree better than they predict, on enough lines, with enough discipline, to overcome the margin. That’s what closing line value tracking is built to measure, and it’s the natural next step from this kind of probability work – exactly what my closing line value guide for MLB outrights walks through in detail. Without the implied probability layer underneath, CLV is just a number. With it, CLV becomes the report card on whether your disagreements are correct.