Two in Three Home Run Hitters Do It Again. Most Doubles Hitters Don’t.
Is a home run a clean sign of power? Ask any fan who the power hitters are and you will get a list of home run totals. So we checked whether that list means anything a year later.
Here is the answer, with its number. We took every regular hitter in back-to-back seasons from 1997 to 2025. Of the men in the top third for home runs, two in three were still there the next year. For doubles, fewer than half. If the numbers were pure luck, one in three would stay.
The surprise is in the doubles. You would think the double and the homer are cousins, two kinds of extra-base hit, both “power.” Across these seasons, a hitter’s home run rate told you almost nothing about his doubles rate. By the end of this, you will know which of the two to trust, and why.
This is the second issue of Box Score Anatomy. Same four questions as always: what is it, how do you get one, what is it a sign of, and what else is a sign of the same thing?
What It Is, and How You Get One
A home run is a ball hit over the fence in fair territory, or, rarely, a ball the batter circles the bases on before anyone can tag him. It scores him and everyone on base.
Notice what it takes. Bat speed and strength, to hit the ball hard. A swing that lifts it, because a hard grounder is a single. Then the things that are not him: the size of the park, the ball that season, and the pitcher who left one over the plate. But most of the recipe is his; that is why it repeats.
Look at the bottom bar. A doubles total is closer to a coin than a skill. Some of that is real; line-drive hitters and fast men hit more of them. But a lot rides on things that move year to year: where the ball lands, how big the outfield is, whether a fielder cuts it off.
A home run total is one of the few numbers you can trust next year. A doubles total mostly is not.
What It Is a Sign Of
Here the arrow is simple. Power causes home runs, so the home run count is a fair window onto power. Statisticians call that a reflective indicator: the hidden thing pushes the number around, so the number repeats when the thing does.
Doubles have more arrows pointing in. Power, yes, but also speed, the park’s walls, and luck. When many causes feed a number, and some of them change every year, the number wobbles. That is the whole difference.
Now notice the runs batted in bar, tied with home runs at 67%. Did the first issue of this series get RBI wrong? No. RBI repeat because the things feeding them repeat too: a hitter keeps his power, and his manager keeps batting him fourth. A number can repeat and still measure his teammates. Repeating is necessary for a good indicator. It is not enough.
What Else Is a Sign of the Same Thing
The best other window is isolated power: slugging percentage minus batting average, the extra bases per at-bat. It moves almost in lockstep with home run rate, because most extra bases come from homers. And strikeouts climb with home runs. The swing that lifts the ball also misses it. That is the price of power, and it shows up in the box score too.
What About 1998?
That was the summer everyone counted home runs. Mark McGwire hit 70 and Sammy Sosa 66. Did the home run race hide better hitters? Mostly, no; the man in front was also the best hitter. McGwire also led baseball in on-base percentage. The homers were honest.
But look one spot over. Barry Bonds hit 37 homers, plus 44 doubles, 130 walks and 28 stolen bases. Craig Biggio led baseball with 51 doubles. Neither one was the story that summer. And Sosa won the National League MVP with only the fifth-best OPS in baseball, carried by 158 runs batted in. The box score had all of it. The headlines picked one column.
What to Take Home
When you judge a hitter, trust the stats that come back. Home runs come back. Walks and strikeouts come back even more. Doubles and batting average come back much less, so a big year in either is partly a lucky one.
And when one number takes over the conversation, the way homers did in 1998, ask what the rest of the box score is saying. Somebody is usually doing something great in a column nobody is reading.
How this was measured, and what it cannot do
Data: every hitter who qualified for the batting title from 1997 to 2025, from the MLB Stats API. We paired each hitter’s season with his next one when he qualified in both, which gives 2,516 pairs. Pairs that touch the 60-game 2020 season are left out. Each stat is per plate appearance; “top third” is set within each season.
Other measures: the season-to-season correlation tells the same story: home run rate 0.75, batting average 0.51, doubles rate 0.33. Within a season, home run rate and doubles rate correlate at 0.10. The 1998 figures were checked against the same feed’s league leaders.
What it cannot tell you: only hitters who kept a full-time job in both years are counted, so slumps that cost a man his job drop out. That flatters every stat a little, though not one more than another. Parks and the ball change over time, and we did not adjust for either.
Argued about something you could not settle? Send it to us · what makes a good question