IS THE DIFFERENCE BETWEEN HITTING WITN NO OUTS & HITTING WITH 2 OUTS INCREASING?

(LONG time no see...life is hectic and fraught on many levels: we can only hope that the chaotic events that continue to plague us here in Amerika will not disrupt or dislodge our so-called National Pastime...)

THIS is a preliminary, high-level first cut at a phenomenon that seems to have escaped attention--which, of course, is one of the main things we do here...

The long table (at right) captures OPS values for hitting with 0, 1, and 2 outs over time: the source (of course) is Forman et soeur (that's Baseball Reference to the rest of you).

Color coding here is as follows: 1) orange-shaded areas show seasons in AL and NL were OPS values with 0 outs meet or exceed .750 (among other things, providing us with a vibrant identification of the long "offensive explosion" from 1993-2009); 

2) green-shaded areas show seasons where OPS values with 0 outs are less than .700 (clustering in the expanded strike zone era of1963-68, and cratering in 1968--but lingering intermittently into the 1980s); 

3) seasons where the OPS values with 1 or 2 outs are higher than the OPS value with 0 out are shown in red type;

4) seasons where the league ratio of 2-out OPS is higher than 100% compared to 0-out OPS are shown in bold type and highlighted in yellow (as will be clear from a full examination of the table, this is extremely rare, occurring in only two seasons--NL 2022 and AL 1984--over the sixty-six years of data);

...and, finally, 5) extremely low league ratios of 2-out OPS values to 0-out OPS values are shown in green type.

WE ordered the data in reverse historical order so that the reader might more easily pinpoint the accelerating occurrence of low 2-out OPS to 0-out OPS ratios. Not all ratios lower than .950 (meaning a greater than 5% drop...) are highlighted, but the first strikingly low ratio occurs in NL 2006, followed by a increasingly frequency of incidence over the most recent twenty years.

In the current decade, there has been some level of retrenchment: the even-numbered full seasons (2022 and 2024, years in which HR/G totals swung downward) returned these ratios more into overall historical alignment, but there is greater year-to-year volatility in the data than ever before.

The average ratio of 2-out OPS to 0-out OPS over the sixty-six seasons is .965 (actually, .965 in the AL, .964 in the NL), meaning that offensive production has had a pervasive decline of 3.5% when hitters hit with 2 outs. To understand the dynamics of this in a systematic way, we'd need to see if there are consistent patterns in sub-details of the data, such as inning-by-inning ratios--is the drop greater in late innings, and is that driving the 21st-century decline in the 2-out to 0-out ratio? (Forman et soeur does not provide data at that level of detail, but we will ask...)

WHAT we can do at this point, though, is show you how the basic data has shifted over time. At left are the decade-by-decade averages for the 2-out to 0-out OPS ratio (along with breakouts for two longer historical ranges: 1960-1989 and 1990-2025). Focus on the "delta 0-d" columns for each league and notice where the decade-level changes occur. While the NL ratio starts to drop in the 1980s, both leagues experience a significant downturn even as offense surges in the 1990s (a reflection of some form of change in hitting strategy, perhaps?). 

The general downturn in offense in the 10s and 20s has not brought the ratio back up to its pre-1990 levels, and we can see in the three-decade breakouts (1960-1989, 1990-2025) that the gap in the 2-out to 0-out OPS ratio has doubled (from 2.5% to around 4.5%).

WE'LL need to look at what team-level data can tell us--whether, for example, the pitching staffs of teas that make the post-season are getting better at getting the third out relative to the league--to see if we can determine why this is happening, but it looks clear that this is not a case of random variation (despite the year-to-year fluctuations). And a more systematic look at the often-ignored "contextual stats" (runs scored and RBI) might reveal something about the change, which might relate to the slow desiccation of the "long-sequence offense" despite the ups and downs of offensive levels.

So we close by quoting that noted philosopher Stephen Stills (pithier and far more of an MVP in his wayward way than our overly-exalted high-school classmate Michael Sandel, ironically lionized for devising a simperingly perverse but ultimately empty form of "communitarian[ist] condescension"...) in the charismatic guitarist's signature moment of confused clarity--"there's something happening here...what is it ain't exactly clear"--as today we watch more of the world burn for no good reason, and look for solace wherever we can find it...even in the wishing well of baseball statistics (where there is more truth to be found than in Sandel's philosophy). 

The hole we have dug ourselves is even deeper than that wishing well, but regaining our (metaphorical) ability to hit with two outs is now as crucial to our life on this planet as it is for the ongoing health of what we still (quaintly) call "the National Pastime"...

NEW TOOLS FOR THE HALL OF FAME: 2B BY AGE-DECADES

AS you can imagine, the Hall of Fame vote results (tomorrow) are trapped in amber somewhere between celebration and disdain, owing to the onerous politics that have evolved in the post-neo sabe era. We are resigned to the fact that Andruw Jones will be in the Hall of Fame via the front door (either this year or next), which is a mistake being perpetrated by the WAR-besotted contingent now ensconced at the center of this decision-making--but that's a matter for another time. 

Instead, let's address another way into measuring career performance that might allow us to escape the prison-house of WAR, and possibly even dovetail into the emerging shift in Hall of Fame voting that is now underway (recently articulated by the less doctrinaire members of the current BB-WAA such as Jayson Stark and Ken Rosenthal). That shift has to do with lowering longevity standards for front-door HOF induction. The rationale is as much a political response as one formed from rigorous analysis (but what else is new?)

For lack of a sexier term, let's call this method the Age-Decade Index.

The method involves evaluation of players at their primary defensive position, which has not always been sufficiently addressed in HOF discussions. (As some of you know already, the WAR approach to positional adjustment is severely flawed, despite the conquer-by-proliferation tactics being ruthlessly deployed by the Tango Love Pie and his followers.)

For purposes of demonstration, we'll use second basemen--for several reasons. First, second base has been a problematic position for the BBWAA and the various semi-feckless committees that have attempted to assemble a sufficient and reasonable group of inductees. Second, we can dovetail the latest committee result--the induction of curmudgeonly keystoner Jeff Kent--into an examination of how a look at age-decades can provide us with greater nuance than WAR or its often-silly offshoot "systems"--especially those devised by deceptively genial egotists like Kent detractor Joe (the Poser) Posnanski. (Space does not permit us to address the massive flaws in his latest agglomeration here; as is almost always the case with Joe, his attempts at "analytical systems" are fatally compromised by his overweening desire to reflect some particularly suppressive subset of "received wisdom.")

Third, and arguably most important, we can use age-decades to bring achievement by decade and by position into alignment in a way that will better define the parameters of the currently evolving changes in HOF voting approaches. (Otherwise, these will remain tainted by WAR.)

SO how does "age-decade" analysis work? The chart at left will be our first guide (but, as Bette Davis said, fasten your seat belts). For each decade (20-29, 30-39), we create age lists based  solely on offensive performance. Because we are neolithic analysts (as opposed to neo- or post-neo sabes), we use OPS+.

Using OPS+, we create ranking lists at each individual age for (in this case) second basemen aged 20-29. To make the list, these folk must have had 400+ PAs in the age-year with an OPS+ of 100 or higher.

We create penalty values for each individual year based on how many second basemen had an OPS+ of 100 or higher. Those penalty values are lower in ages like 20, 21, 22 because fewer second basemen (and few players in general) qualify for the list. The penalty value is higher in ages 26, 27, 28, 29 because more players make the list.

There is some nuance in assigning these values: a player with an OPS+ > 100 but with <400 PAs will generally have the specific penalty value for that age cut in half. 

Once all rankings & penalty values have been assigned, we sum the values for each 2B and average them. In this configuration, lower is better, so we flip the value on to a scale where 100 = perfect. 

Then we sort the list in descending order, and (just because we can) we calculate the number of years in which they make the list, and the number of years they crack the Top Ten.

After all that, we discover that Eddie Collins was the greatest second baseman during the age 20-29 decade, followed (not so closely, as it turns out) by Rogers Hornsby (not always a 2B) and Nap Lajoie.

This is not so surprising, of course, because most systems would generate similar results. (Sometimes greatness is truly obvious...)

BUT who knew that big little man Jose Altuve (still in pursuit of 3000 hits) is right on Lajoie's heels? And who the heck are those guys named Cupid Childs and Larry Doyle? Their Age-Decade Index values for their twenties are mighty good: if they both hadn't played prior to 1920 and buried in the mists of time, they might already be in the HOF along with several (Bobby Grich, Lou Whitaker) whose absence remains haunting for a certain coterie (identified during Christina Kahrl's "cheeky phase" as "armchair analysts"). 

Note, BTW, that actual HOF inductees are shown above in bold type. We kept a sample of also-rans displayed on the list to give you an idea of how deep it goes.  And we showed you how HOFers who switched away from second base (Rod Carew and Paul Molitor) were faring before they moved to another defensive position.

Those also-rans are also useful in order to see how the math works, as evidenced by the adjacent entries for recidivist Twin Edouard Julien and Federal League flash-in-the-pan Duke Kenworthy. (Julien hasn't come close to his fabulous rookie season in 2023 (age 24). but the method isn't ready to penalize him for later-age seasons that haven't occurred yet; Kenworthy dominated a tetchy major league in 1914-15 (age 27-28) but could never pass muster in the real big leagues, It seems appropriate that they are lumped here together...)

And many of you 19th century aficionados will know that the mysterious fellow named "William Robinson" is actually the early walkman better known as "Yank."

But another Robinson--Jackie--does not show up well here, owing to the fact that he couldn't play in MLB until he was 28 (and in that year he was forced to play first base). We'll see how he fares in the 30-39 decade. 

AND then there's Jeff Kent. How did a guy with such a middling offensive performance in his twenties possibly wind up inducted into the HOF (even by a "vets committee")?

But before we do that, we want to sell you some insurance...or, should we say, we want to show you BBB's version of an actuarial table--a tool that should be more front-and-center in baseball analysis for many reasons. This particular actuarial table (at right) reveals some things not generally acknowledged in terms of second basemen and their career survival rate.

We have here the total number of second basemen in baseball history with 400+ PAs in all the age-years from 20 to 40. If we graphed the "Total" data we'd see that it is a front-loaded bell curve with a steep early ascent and a long decline. The general finding is that there are nearly twice as many 400+ PA seasons at second base during the 20-29 decade as is the case for 30-39. The number of them that are above 100 OPS+ (and above 120 OPS+) start declining at age 23 as teams settle for good glove men when the supply of good young hitters at the position lag (in part because some of the better hitters get moved to other positions). The top hitters at 2B (120 OPS+) tend to survive longer relative to the number of 2Bs with 400+ PAs, which declines rapidly starting at age 32.

But these survivors are even rarer than the more general actuarial data. Age 20-25 2Bs with 120+ OPS radically outnumber their counterparts at age 35-40 (79 for age 20-25, just 24 for 35-40). Such a performance really is exceedingly rare--and if it is something that gets repeated multiple times during that age range, it is deserving of weighted consideration with respect to the Hall of Fame

So, to add nuance to the arguments now being made by Stark and Rosenthal, it isn't just the peak that should get greater emphasis, but also the location of the peak within specific age ranges.

AND that brings us back to Kent. The data for 2Bs in the age 30-39 range (table shown at left) provides astonishing confirmation that it is the Curmudgeonly One who scores best in a year-by-year anatomization. (Just barely, though, over Joe Morgan...)

And this is especially the case in the age 35-39 range, where such performance levels are almost exclusively limited to HOFers.

The all-time greats (Collins, Hornsby, Morgan, Lajoie, Gehringer) are still significantly greater than Kent, of course, due to their early achievements. 

But here is where Kent leapfrogs into the Hall, in defiance of WAR and the even more distorted, jackleg method coughed up by Joe the Poser (apparently due to a tantrum over the latest HOF rejections of his favorites). At best, the tool he devised should be in a group of measures with at best a tertiary level of weighting.

WIT the Age-Decade Index, we also see that the criminally overlooked duo of Grich and Whitaker score well here, confirming their worthiness for enshrinement in Cooperstown. We see the short-term dominance of Jackie Robinson, too. Finally, the method also validates claims to be made on behalf of Robinson Cano (don't you know)...and gives a boost to Chase Utley, too.

We think the method needs a little more calibration before the 20-29 and 30-39 ranges can be combined to form a fully unified career evaluation--in particular, the weighting between the two decades requires some additional refinement. We'll work on that and present it here in a subsequent post...stay tuned!