A win is not worth the same at every price.

The missing half of win rate

A record of seven wins and three losses sounds strong. It is incomplete until the prices are known. A favorite priced at -250 returns only $40 of profit on a winning $100 stake, while a loss costs the full $100. Seven wins at that price earn $280; three losses cost $300. The result is a $20 loss despite a 70% win rate.

Break-even probability

Break-even probability is the win rate needed for the price to produce neither profit nor loss before any other costs. For negative American odds, divide the absolute odds by the absolute odds plus 100. For positive odds, divide 100 by the odds plus 100.

At -200, the break-even rate is 200 ÷ 300, or 66.7%. At +150, it is 100 ÷ 250, or 40%. Those thresholds make win percentages comparable to the prices that produced them.

Two 40% records can mean different things

Four wins and six losses at +150 break even: four $150 profits offset six $100 losses. The same 40% record at +100 loses $200. The win rate is identical; the economic result is not.

What to inspect in a published record

Look for the exact price recorded with each selection, the time it was published, the grading rule, net profit, total amount risked, and sample size. If a performance claim gives only wins and losses, you cannot reconstruct its return without the missing prices.

ClearInning keeps win rate visible because it is useful descriptive information. It is evaluated beside published-price ROI, never as a substitute for it. Read the ROI guide for the full ledger calculation.

AI-assisted review does not change the math

AI-assisted checks may help identify inconsistent arithmetic or unclear wording. The formulas, recorded prices, and result ledger remain auditable inputs. AI review does not improve a price after publication and does not guarantee an outcome.