Every game was priced from real bookmaker moneylines pulled from the historical odds archive, then de-vigged so the two sides add to 100%. That de-vigged number is the market's honest win probability. A factor only counts as an edge if games matching it beat that probability by more than chance would explain. Every game split by home/away and favourite/underdog using the real closing moneyline. Home underdogs isolated, then their real win rate compared with the summed de-vigged probability across the same games. Sub-buckets by how big an underdog they were, to check for an effect hiding at one end of the range.

edge lab · factor writeup · 6 min read

Home Underdogs: Is the Crowd Worth Anything the Market Missed?

Q: What the numbers said

Sample: 5,295 games; Window: 2006–2025; Signal: z < 1.17 No exploitable deviation, overall or in any price band. Home underdogs win often enough to feel memorable and not one game more often than they are priced to.

Q: What it means for your picks

This matters twice in a survivor pool. First, do not talk yourself into a home dog because the story is good. Second — and this is the useful half — do not panic when your pick is a road favourite. Home-field advantage is real, it is worth roughly two to two and a half points, and the price you are looking at already contains all of it. The practical version: Trust the de-vigged number. It is the single strongest signal we found, and every attempt in this lab to beat it failed. Live prices with full book-by-book depth are on the odds board; the rest of the register is in the Edge Lab.

Home dogs are the classic contrarian bet. Tested across 5,295 games, they win exactly as often as their price says they will.

The home underdog is a folk hero of sports betting: the disrespected team, the loud stadium, the number that looks a point too generous. It is also one of the most-bet situations in the sport, which is a clue about how it turned out.

What we tested

The hypothesis: teams that are underdogs at home are systematically undervalued and win more often than the market implies.

How we tested it

Every game was priced from real bookmaker moneylines pulled from the historical odds archive, then de-vigged so the two sides add to 100%. That de-vigged number is the market's honest win probability. A factor only counts as an edge if games matching it beat that probability by more than chance would explain.

Every game split by home/away and favourite/underdog using the real closing moneyline.
Home underdogs isolated, then their real win rate compared with the summed de-vigged probability across the same games.
Sub-buckets by how big an underdog they were, to check for an effect hiding at one end of the range.

What the numbers said

Sample: 5,295 games
Window: 2006–2025
Signal: z < 1.17

No exploitable deviation, overall or in any price band. Home underdogs win often enough to feel memorable and not one game more often than they are priced to.

What it means for your picks

This matters twice in a survivor pool. First, do not talk yourself into a home dog because the story is good. Second — and this is the useful half — do not panic when your pick is a road favourite. Home-field advantage is real, it is worth roughly two to two and a half points, and the price you are looking at already contains all of it.

The practical version

Trust the de-vigged number. It is the single strongest signal we found, and every attempt in this lab to beat it failed.

Live prices with full book-by-book depth are on the odds board; the rest of the register is in the Edge Lab.

More: back to the research register · methodology · guides