What the heck is the expected wins metric?

Look: it’s the Pythagorean expectation boiled down to a single number that predicts how many games a team should win based on runs scored versus runs allowed.

Crunching the numbers

Here is the deal: you take total runs scored (RS), raise it to the power of 1.83 (the magic exponent for MLB), do the same with runs allowed (RA), then plug into the formula Wins = Games × RS¹·⁸³/(RS¹·⁸³ + RA¹·⁸³). Simple, brutal, effective.

Why 1.83?

Because the data tells us that exponent fits the league’s run distribution like a glove; lower than 2 because baseball isn’t a pure power-law sport, higher than 1 because scoring isn’t linear.

Applying it to a season

Take a team that scored 750 runs and gave up 680. Compute 750¹·⁸³ ≈ 3.2 × 10⁵, 680¹·⁸³ ≈ 2.5 × 10⁵. Plug into a 162-game schedule: Wins ≈ 162 × 3.2/(3.2 + 2.5) ≈ 95 wins. That’s the expected win total, regardless of luck.

Spotting over- and under-performers

When a team’s actual win total deviates wildly from this projection, you’ve got a regression candidate. If they’re 10 wins above expectation, they’re either lucky or have hidden talent that’s not reflected in run totals.

Common pitfalls

Don’t feed the formula raw box scores without context; park factors, bullpen usage, and defensive shifts can skew RS and RA. Also, don’t treat the exponent as a static constant — some seasons it drifts toward 2, others toward 1.7.

Real-world tweak

Advanced analysts often replace raw runs with weighted runs created (wRC) to smooth out park effects, then re-apply the same exponent. The result? A tighter fit to actual win-loss records.

Quick actionable tip

Grab the latest team RS and RA, plug them into the formula, compare the output to current standings, and flag any club that’s more than five wins off the mark — those are the teams you should be watching for a swing in performance.

For a deeper dive, check out this baseball expected wins formula article.