In baseball, single-run leads are relatively durable because runs tend to come in bunches and the default outcome is not to score at all.

The claim has three parts: (a) one-run leads in baseball are relatively durable, (b) runs tend to come in bunches (scoring is clustered or overdispersed across innings), and (c) the most common outcome of an inning or plate appearance is that no run scores. True: sabermetric or statistical evidence (e.g., win expectancy tables, run-distribution data) supports all three, and the causal link (bunching plus frequent scoreless innings make small leads hold more often than a uniform scoring model would predict) is consistent with standard analysis. Mostly true: all three facts hold but the causal explanation is only partly supported, or 'relatively durable' is somewhat overstated compared with the benchmark the claim implies (other sports or a uniform-scoring model). Mixed: one of the three parts is contradicted or clearly unsupported while the others hold. Mostly false: the data show one-run leads are not notably durable, or scoring is not clustered, while the claim rests on some true element (e.g., most innings are scoreless). False: the evidence contradicts the core assertions, for example most innings do produce runs, or one-run leads are less durable than comparable benchmarks.

Set before the research began.

1. E1 Run Distribution Using the Negative Binomial Distribution FanGraphs Community · 9 September 2014

   “So if you are trying to describe why 73% of all MLB innings are scoreless to a friend over a beer, either will work.”

2. E2 Beyond runs expectancy SAGE Publications (Journal of Sports Analytics) · 4 September 2015

   “Generally, the Poisson fit understates the actual variability in runs scored.”

3. E3 leverage – Statistically Speaking Archive StatSpeak (Pizza Cutter) · 16 February 2008

   “One run lead: 15.2 times (3.69 blown saves, 24.3% fail rate)”

4. E4 Baseball Baselining: How Likely Is a Comeback? FanGraphs · 9 August 2023

   “By the start of the ninth, the odds of a comeback are 15% in a one-run game, 5% in a two-run game, and 2.5% in a three-run game, which means you can almost ignore three-run games and treat them as over.”

E1 reports that about 73% of MLB innings are scoreless, which supports the claim that not scoring is the default outcome. E2, a paper in the Journal of Sports Analytics, finds that a Poisson model understates the variability in runs scored, which fits the idea that runs come in bunches. E4 says teams come back from one run down at the start of the ninth only about 15% of the time, so a one-run lead holds about 85% of the time at that point; E3 gives a 24.3% fail rate for one-run leads. None of the evidence compares baseball with other sports or tests the causal link the claim draws, so 'relatively durable' and the explanation are only partly supported.