I have mixed opinions about the designated hitter rule. I don't really mind it, as it gives players like Jim Thome and Edgar Martinez the ability to contribute for longer or more often than they would have otherwise, most due to injury risk or age. At the same time, it's an impure hack, particularly the way it's implemented. The DH can only hit for the pitcher, for instance. And it removes tactical tradeoffs that would otherwise matter. And some pitchers can actually hit, too.
So the current situation works well for me, with about half of baseball using it and the other half playing the more pure, tacitcal version of the sport. That's why I chose the 81 games with DH, 81 without hypothetical season. I think it well reflects the effect of the DH on the game. Naturally, the DH is a going to be one of the biggest bats on the team, so I have the added wrinkle of what to do with him when the DH isn't allowed.
There's also the question of whether the DH must have had significant time actually being a DH. That's not an unreasonable requirement, but given that it's a new rule only used in one league and that I'm putting together "the greatest possible baseball team", I decided not to use it.
Selection Ted Williams
Ted Williams is the greatest batter of all time. Admittedly, it's a tight race at the top, with Babe Ruth and Barry Bonds being the most notable competition. Throw in Lou Gehrig, Rogers Hornsby, Mickey Mantle and Ty Cobb, and I think you have most of the major candidates.*
*This is a wonderful and multilayered argument, and one can make good cases for most of these players and a fair number more: Musial, Mays, McGwire, Pujols, Frank Robinson... even the underrated Dick Allen.
My case for Williams over the competition is thus:
Williams is either 1 or 2 all time in the important batting stats: 1st in OBP and 2nd in SLG, wOBA, wRC+ and OPS+. The only player ahead of Williams on any of these lists is some guy named George Herman Ruth.
Independent of context*, the most important batting skill -- the one most important to scoring runs -- is the one Williams leads everyone in: on base percentage. This is because not only does getting on base enable the batter to score, it means he didn't get out. I'll have to sit down and actually do the math sometime, but as OBP goes to one, runs scored per game, inning and out all go to positive infinity. And a high OBP drives up the chances that follow batters will 1) get addition plate appearances and 2) have runners do drive in. A player with a high OBP makes the entire offense more potent. (As does a player with a higher, say, slugging, but the effects aren't as great).
Williams played in a more difficult era than Ruth in particular. While the normalized stats account for some of the difference, segregation in particular casts some doubts on the validity of Ruth's numbers. I'm sure he still would've been brilliant, but excluding black players both decreased the quality of the pitching (and perhaps defenses) Ruth faced and lowered the average hitting stats which Ruth's numbers are normalized against. It's not a large difference, but it eats into Ruth's small advantage.
Also, the majority of Williams career occured during the post-integration, pre-expansion era, where the average player quality was boosted compared to earlier eras because of integration and boosted compared to later eras because expansion hadn't diluted the average quality of the opposition.
Era arguments can be used against Williams, too, of course. Ruth suffers (very slightly) from plate appearences made during the deadball era and old style home run rules where balls that went foul after leaving the park were still foul. Williams's early seasons were also during the segregated era. And Williams faced few Latino and no Japanese players. In general, later eras have larger populations from which to draw players and so the top 400 players in the world are even farther above the mean player than they were in the past. Modern technology, including game advances, health advances and information technology like video also improve the performance of the average major leaguer (and make era to era comparisons harder). There's also the question of what would the past greats done in the modern context? Pie Traynor with a modern glove, Dizzy Dean with a modern pitching coach and medical staff, Koufax with modern surgery and a two seamer, etc, etc.
- William's lost much of his peak to military service and two different wars. Across 1941, 1942, 1946 and 1947, Williams averaged .360/.506/.669/214 (avg/obp/slg/wRC) in 2642 plate appearances at 660.5 PAs a season. Give Williams credit for 660 PAs for 1943, 1944 and 1945 at that rate, and his career line bumps up to .347/.486/.640/193, which roughly halves the distance in wRC+ between he and Ruth's .342/.474/.690/197. This improvement also sets him farther ahead of batters who played in later, presumably more difficult, eras.
A similar excercise can be done for William's Korean service, but it's less clear which years to choose for the reference points. His service also exacted a cost in training time, travel, etc, that's harder to account for.
*Of course, context matters. And I don't mean protection, as that generally has at most a rather small effect. Nor am I addressing the league context, though that is important. Rather, I mean that the batter's teammates matter. On a high obp team, say of Ted William's clones, an individual's OBP matters more. As his teammates are less likely to get out, he'll score eventually, even it takes four singles or walks.
On a low OBP team, power, particularly home run power, matters more. On a team of Ozzie Guillen's, where there's rarely anyone on base when our hitter comes up and the players behind him are almost pathologically unable to advance him, merely getting on base doesn't help much. However, home runs (guaranteed run) and extra base hits followed by productive outs (double, advanced to third on ground out, sacrifice fly) can still generate runs.
So, when choosing between Williams and Ruth (accepting the validity of their raw stats), the better choice really depends on their teammates. In other words, "the greatest ever" question, in team sports in particular, is generally ill defined.
For his career, Williams hit .344/.482/.634/189 and netted 139.8 fWAR. Peak doesn't really fit Williams, "mountain range" might be a better choice. When choosing between his best years, though, his famed 1941 season (.406/.553/.735/221 for 11.9 fWAR) and his post integration masterpiece in 1953 (.388/.526/.731/223 for 10.3 fWAR) are my favorites.
So, even if I haven't convinced you that Theodore Samuel Williams is the greatest batter of all time, you ought to agree he's a top 2, top 3 sort of guy. And since I'm picking a team here, you'll probably see your top guy too.
Column A: Your pick for "Greatest batter of all time" not otherwise on your roster.
Column B: Great batters who actually spent time at designated hitter.
The Big Hurt is the most potent player to spend a significant fraction of his play time at DH, with a career .301/.419/.555/154 line for 76.2 fWAR. He has a legitimate case for best right handed hitter of all time. Thomas's best year was 1994, hitting a Williamseque .353/.487/.729/204 for 7.3 fWAR in just 113 games due to the strike shortened season.
DHs become DHs for a variety of reasons. Martinez, a decent enough fielder at third base, moved to DH due to lingering effects of a hamstring injury in 1992. Taking him gives one more infield flexibility and a still fantastic bat. The right handed Martinez owns .312/.418/.515/148 career line for 69.9 fWAR with his best year probably being 1995: .356/.479/.628/184.
Martinez is notable for not seeing signficant play time until he was 27. It's hard to say what playing earlier would've done to his career rate stats, as batter peaks often occur near 27, but his counting stats would probably put him solidly in the Hall of Fame.
Some players earn immortality by being memorable. Some earn it by being great. Thome can claim those, but as far as anyone can tell, he might actually be immortal, as he's still a good hitter in his age 42 year. Thome's ongoing 22 year career includes a peak season in 2002, where he hit .304/.445/.677/188 mostly at first base.. Here, Thome provides a left handed option for the DH slot with actual DH experience, in addition to his time at 1st and 3rd, hitting .277/.403/.556/145 for his career so far.