Looking Back at the 2018 Postseason

The 2018 baseball season has ended, after a postseason that, in retrospect, felt like a coronation parade. The Boston Red Sox dominated the regular season with 108 victories and did not lose more than one game in any of their three postseason series. After three years of teams breaking decades-long droughts (one over a century long) to win the title, Boston won its fourth in 15 years. For the historic excellence of this year’s squad, and for its lengthening pattern of World Series triumphs, the Red Sox can claim to be the Team of the Millennium.

I’ll be looking back at the postseason today, just as the title promises. As has become my annual custom, I’ll be gauging the excitement of the games rather than the team and player narratives. My method for doing this is the Win Percentage Sum (WPS) Index. Back in 2012 and ’13, I did daily reports on postseason games, supported by WPS. When THT changed its format in early 2014, I shifted to a wrap-up after the postseason. This is the fifth of those retrospectives.

The next several paragraphs give a summary of WPS methodology, though you can refer to my original articles from August 2012 for a fuller explanation. As always, I will first recognize THT’s Editor Emeritus Dave Studeman for creation of the original game excitement metric. When I conceived my system, I did not know of his similar work, published in the 2007 Hardball Times Annual. Dave graciously gave me leave to continue using WPS, and I gladly give credit where it is due.

WPS is calculated with a base score that adds together the change in game Win Expectancy created by every play, meaning every continuous action that changes the base-out state of the game. If a player’s pop-out lowers his teams chance of winning by five percentage points, or his single raises the chance by five percentage points, that’s a score of five either way.

(Last year, I made a slight alteration to this method. When intentional walks were made automatic, with no pitches required to be thrown, I stopped adding their Win Expectancy numbers to the WPS score. My logic is that if there is no action on the field, there is no excitement gained from game play. I continued that policy this postseason.)

To the base score I make two additions, meant to reflect fans’ subjective enjoyment of the high points of a game. I add the three plays with the highest Win Expectancy changes to the base, then add the value of the game’s final play. These reflect, respectively, the highlights of the game and how much the game was in doubt at the very end.

If a big walk-off play is one of the top three, that means it counts three times: in the base, in the highlights, and as the final play. This is intentional. Fans love walk-off wins, especially those swinging a game from defeat to victory, so the WPS system favors them. As an example: Kirk Gibson’s famed walk-off homer in Game One of the 1988 World Series had a Win Percentage Added of 87 percent. In the WPS system, it thus added 261 points to the score of the game, turning a merely good game into an all-time great baseball memory.

The median WPS score of a game is roughly 300, with wide variations running from below 100 to over 1,000. Under 200 denotes a snoozer, while 500 points is my criterion for a “great” game. (The above example, which would have ended in the low 400s if Gibson struck out, came in at 632.5.) The mean of WPS scores is somewhat higher than the median, because there is more numerical room for a game to be great than to be boring.

The system has blind spots. Outstanding individual performances or milestones count for nothing, except by how they affect the game’s competitiveness. No-hitters and perfect games are hardest-hit. Since they usually lack see-saw movement between offense and defense, or between the competing teams, their scores generally are flat. Best-play and final-play bonuses can capture some subjective reactions, but others are beyond a system with any pretension to simplicity.

Another, more equivocal, feature of the system is that it believes more baseball is always better. WPS points can only accumulate with added plays, never diminish. Thus, extra-inning games have a structural advantage, growing greater the longer the game goes. That came into play a couple times this October.

Along with the WPS scores, I will use a second method to judge this year’s playoff games: the CPS system, which I introduced last year. Championship Percentage Sum, or CPS, combined the WPS scores with Championship Leverage Index, the brainchild of Sky Andrecheck. Andrecheck’s CLI measures postseason games by how much they move their teams toward, or away from, a World Series victory.

As examples: Game Seven of the World Series is worth 1.0 Championships Added. Game Six of the Series is worth 0.5, the same as Game Seven of a League Championship Series. Game Five of a League Division Series is worth 0.25 ChampAdded, while Games One and Two are worth 0.09375 apiece. Here is the full breakdown of Championship Leverage Indices.

The CLI works very similarly to the in-game Leverage Index. The closer a series stays, the higher the leverage moves. A high WPS-scoring game late in a series will send the CPS soaring, the way a three-run homer breaking a tie in the eighth would score much higher than the same play in the third. This way, a blockbuster Game Seven is considered to make a bigger impression on fans than if it happened in Game Two. This is certainly true, though whether it occurs in the ratios the CLI uses is debatable.

I believe this is an improved way of gauging the excitement of postseason games, but reasonable fans could disagree. If nothing else, consider it a second eye on the action, providing greater depth to our view of events.

The Appetizers

Before we reached the actual postseason, we got a dry-run doubleheader. The National League Central and West divisions both ended in ties that had to be broken in order to crown champions and send the losers to the Wild Card knockout game. Given the importance of the games, even in what amounted to a double-elimination format, I put both contests through the WPS machine.

The Milwaukee Brewers and Chicago Cubs played first, and produced a decent game. Its 313.0 WPS score was a smidgen above average, with the tightness of the score for most of the game balanced by a relative dearth of scoring threats. (Seven half-innings went 1-2-3, which are the dullest results the system can have.) Also helping fan interest was that the game concluded in three hours, six minutes. Had this been an actual playoff game, it would have been the second-shortest of 2018.

The Colorado Rockies and Los Angeles Dodgers were a steep drop-off. Walker Buehler smothered the Rockies while he was pitching, and L.A. methodically built up a five-run lead in the middle innings to turn the late frames into a WPS desert. A two-run Colorado comeback in the ninth scarcely moved the needle, and the game ended with a flat 194.1 score.

Tie-breaker games, or playoffs as they were once known, are historically fertile ground for great baseball. Of the 22 such games played in major league history, from the 1946 National League playoff series to 2018’s, five have scored in 500-plus “great game” territory. Two others, the final game of the 1962 Giants-Dodgers series and the 1978 Yankees-Red Sox “Bucky Dent” game, were near-misses in the 490s. The highest scoring was the 2009 Twins-Tigers epic, at an amazing 916.6. This year’s Game 163s sadly did not add to that store of thrilling baseball.

The Wild Card games did better in that respect. Well, at least, one of them did.

The Rockies-Cubs game was a pitchers’ duel, and those generally underperform in WPS’ eyes. It made up for that by being close, with neither team ever ahead by more than one run, and by going deep into extra innings, which piles up points. The two highest rating innings were the 11th, when both teams threatened but neither scored, and the 13th, when Tony Wolters pushed Colorado ahead for good. The game registered a 628.4, making it the second-best Wild Card game ever, behind 2014’s epic 12-inning tussle between the Royals and A’s.

The American League counterpart was rather more forgettable. The Yankees went up 2-0 on the A’s two batters into their first inning and didn’t let Oakland score until they had pushed the lead to 6-0 in the sixth. Oakland threats in the fourth and fifth provided the most excitement the game provided, but they didn’t close the gap and couldn’t lift the game past a WPS score of 202.5. This is not historically terrible: Out of 13 other Wild Card games ever played, four have had lower WPS scores. The Wild Card round, especially in its earlier years, has been more of a snoozer than you’d think elimination games would be.

Main Course

Speaking of snoozers, we now have to look at the divisional round.

Professional commentators lamented how boring this year’s Division Series were, and WPS backs them all the way. Of the 100 Division Series that have now been played (1995 to 2018, plus the 1981 strike season), this year’s best was Dodgers-Braves, which scored a lame 66th out of those 100. Milwaukee-Colorado was 74th-best, Boston-New York 87th, and Houston-Cleveland finished 91st. Add up the WPS scores for all four series, and the 2018 divisional round comes out as the most boring of the 25 ever played. (The total is 3767.4 points, undercutting the 2006 LDS round at 3944.0)

CPS took a somewhat more lenient view, in one case. While the other series stayed within three places in the rankings, Red Sox-Yankees got boosted from 87th to 66th. The overall effect, though, didn’t move the LDS round out of the cellar. The range of aggregate CPS scores for the LDS went from a high of 872.5 (for 2012) to a low of 357.725 (for 2006), with 2007 a mere 0.025 better. 2018 came in at 347.2, taking over as the least exciting ever.

What made this year’s divisional series so awful? Short series were crucial, though not fully decisive. 2018 went 14 total games, two over the minimum, but four previous LDS rounds had done the same, and 2007 and 2009 went just one game past the minimum. No games reached the 500-point threshold of “great,” or even hit the 450-point mark, while four contests ended below 200 WPS points, my unofficial cutoff for “snoozers.” WPS was bound to hate this round.

Why couldn’t CPS find a reason to disagree? The answer, as fits with a metric based on leverage, is probably timing. The timing of which teams won which games, and when exciting or tedious games were played, added up to a big goose egg this year.

Three of the four series went 2-0, leaving the losers on the edge of elimination and lowering the leverage of Game Three, two of which produced sweeps. The third of those series, Los Angeles-Atlanta, had its best WPS score in Game Three, when Atlanta clawed back a win—but as this came in the game with the lowest Championship Leverage, its effect on CPS was thus muffled.

The series that did go 1-1, Boston-New York, got a bump in the CPS standings for the raised leverage of its third and fourth games. The 16-1 rout in Game Three wasted much of that higher leverage, and an average-ish Game Four didn’t make up the lost ground. The series ended up with the worst WPS score ever out of the 65 divisional series that didn’t end in a sweep and fifth-worst of the 65 for CPS.

There were a couple of highlights to grab our attention. Colorado’s ninth-inning rally to tie Milwaukee 2-2 in Game One was a spine-tingler. The Rockies complied 93.6 WPS points for that comeback, the highest half-inning score of the postseason. The Yankees, down 4-1 in Game Four with three outs to live, nearly scraped together a season-extending rally against Boston closer Craig Kimbrel. Their ninth inning tallied 79.3 points, as their Win Expectancy rose from 4.4 percent, crested at 33.6 percent, then, as Eduardo Núñez’s throw beat Gleyber Torres to first, fell to nothing.

Those highlights aside, the 2018 LDS round was, frankly, awful. The League Championship Series round didn’t have to be an improvement, but thankfully it was, and a big one.

The five-game Boston-Houston showdown came in 37th out of 66 League Championship Series by WPS, just under average. Of the 18 Championship Series that have gone five (in the best-of-seven era), it came a good fifth. The Los Angeles-Milwaukee series scored 13th out of the 66 by WPS; it was in the middle of seven-game championship rounds, ninth of 17. Combined, the two made for a good LCS round and broke a string of three years with an absolute turkey dragging down the LCS.

CPS didn’t have a much different opinion of the Red Sox and Astros, putting it one spot lower at 38th. Among five-game Championship Series, it scored the series third out of 18. (The CPS system favors longer series more than WPS does, which is why this one got better among fives but a bit worse overall.) The Dodgers and Brewers got a big boost from CPS, rising to fifth overall and landing among the great League Championship Series of all time. This is actually a bit surprising, and I will look at why a bit later.

The ALCS was strong most of its curtailed way, with only the final game failing to finish above the WPS median in excitement. The gem of the series was Game Four, a back-and-forth affair that toted up 550.4 points without even going to extra innings. The game is close to a textbook example of how to please the WPS system, and thus hopefully of how to please fans, so I’ll give it a closer look.

One of the strongest reasons for the game’s excellence—and one of the easiest to miss—is that there wasn’t a single 1-2-3 inning all day. Every frame had at least one baserunner; something was always happening, or at least potentially happening. The path to three outs was always up-and-down, covering more ground, accruing more points. The lowest-scoring inning, the second, was worth 29.7 WPS points, which spread across a whole game (plus the best play/final play bonuses) would push the score over the median of 300 points. There was effectively no inning below average in excitement the entire game.

Next, the answer to the question “Who’s winning?” has to change at least once in a game but need not change more than that. In this game, the answer changed six times. (The three possibilities being, “The Red Sox,” “The Astros,” and “No one: it’s tied.”) Again, that shows a weaving path to the end that raises excitement. In two other half-innings, the existing lead was narrowed without being overcome (both times by Houston), which isn’t as strong at raising excitement, but it never hurts.

Finally, the Astros kept scrapping in their last licks, getting the tying run aboard with one out and the potential winning run on base with two outs. Granted, it would have been much more exciting to pull off the walk-off win, or at least force extra innings, but that’s really the only area where this game fell short. (Aside from taking over four and a half hours, but we’ll get to that.) Still, the Astros and Red Sox packed as much excitement into nine innings as any fan could hope for.

Earlier games in their series benefitted from staying close until late. The first three contests were all one-run affairs at the seventh-inning stretch, though they ended with winning margins of five, two, and six runs. Only Game Five fell short, with Boston jumping ahead 4-0 in the top of the sixth and the Astros never getting the tying run to the plate. That made it the lone below-average game of a short series.

The Dodgers and Brewers had five of their seven games finish below the 300 median, though four of the five were better than 250. This meant one or two very good games could make up the deficit, and they got those two. Game Two had L.A. making up a three-run deficit with two in the seventh (the rally then dying dramatically with the bases loaded) and two in the eighth, as Justin Turner’s two-run shot recalled some of his glory from last October.

Then came Game Four. Just like the Rockies-Cubs Wild Card match, it ended regulation knotted at 1-1 and went 13 innings. The games finished within a few points of each other, well within any reasonable margin of error. Which one you think was the most exciting game of the postseason will thus depend on personal taste, including how tolerant you are of winning-run scorer Manny Machado and his 10th-inning sideswiping of Jesús Aguilar at first base. And no, the WPS system takes no notice of the benches clearing.

(I’m not forgetting that other extra-inning game I just left out of the “most exciting” conversation. We’ll talk about it, soon.)

The series gets its CPS boost by something like process of elimination. Other seven-game Championship Series either had a bunch of duds, such as the 1992 NLCS with four games under 160 (!) that undermined the Sid Bream Game Seven, or had an outright stinker in Game Seven, like the 1986 ALCS that went from Dave Henderson’s glory in Game Five to a concluding anticlimax. This year’s NLCS, even without a noteworthy Game Seven, outpaced most of its competition by never outright stumbling.

The World Series left many people stumbling, in a daze—mostly after they stayed up watching Game Three.

WPS liked the Series this year, ranking it 30th out of the 114 Fall Classics ever played. The epic Game Three scored almost the same as three average games, which in effect made a five-game series seem like it went the limit. CPS wasn’t as impressed, rating the series 65th out of 114, a drop almost a third of the way down the leaderboard.

This is where CPS comes into its own. Both WPS and CPS are meant to simulate subjective fan reactions to games and series. My subjective reaction, cleansed, I hope, of my own rooting interests (We fans care whether our favorite or most-hated teams are in the World Series. WPS and CPS don’t. They are meant to measure subjectivity objectively) was that the World Series was much closer to dullish-middling than to solidly above average. CPS, I believe, got it right.

The roots of the disagreement are mainly two-fold: the short series, and the very long Game Three. In fact, the former ties into CPS’s view of the latter, so let us contend with that historic game.

An 873.8 WPS score for a game is of course dizzyingly high. It stands as the third-highest-rated World Series game ever, behind Game Six in 2011 and Game Three in 2005. Those examples give some clue as to CPS’s attitude toward the game. The Cardinals-Rangers epic, coming late in a series that went all the way, is remembered as an all-time classic, which it was. The White Sox-Astros epic, coming at the 2-0 stage of a Chicago sweep, doesn’t have nearly the same standing.

While much of the 2018 Game Three’s WPS score stems from its tremendous length, this can be overstated. The playoffs had an 18-inning game just four years ago, Game Two of the 2014 Giants-Nationals NLDS. That game scored 686.8 points, well behind this year’s marathon. Pure length matters, but it isn’t enough to make an all-time classic.

One thing this year’s game got right was the 13th inning. Boston pushed across a run in the top half and threatened more before being retired. L.A. had to pull a Houdini on what should have been the final out of the game, getting Max Muncy home on Ian Kinsler’s error to re-tie the game. (Muncy would contribute again, with the walk-off home run.) A back-and-forth like that in extras (or the ninth) is gold for excitement. The full 13th scored 127.2 WPS points, the highest total of any inning in the postseason.

That said, Game Three was lacking in certain exciting elements. Scoring was one: The game was 1-1 after nine (just like the Giants-Nationals game). Those early runs didn’t come from rallies but from solo homers, meaning excitement didn’t build, it just happened. The game was chock-a-block with 1-2-3 innings, and many other times when runners did get on, it was with two outs and the inning ended right afterward.

The game could not compile in 18 innings as much WPS as Game Six in 2011 amassed in 11. As I’ve noted in the past, 2011 gave us a perfect storm of a game. Expecting another game to match it is to burden it with unrealistic expectations. Still, much of the WPS score came from great length, and in practical terms this was no blessing.

This was perhaps a game for superfans, not the casual audience. Muncy’s game-winner came at 3:30 Eastern time, which was half past midnight in the West. Pacific Coast dwellers, accustomed to early bedtimes after playoff games, got a taste of what the East Coast increasingly has gone through. The generally suppressed offense in the extra innings didn’t make the time crackle with tension, either.

“It may be a great historical moment,” one friend wrote to me, “but golly it was no fun to watch.” This will likely be the verdict of history, and not only because of the record-setting length. The game came with Boston up 2-0, dropping the Championship Leverage Index. While it felt at the time like L.A.’s win might have turned the momentum of the series, the Dodgers wound up losing the next two. The longest postseason game ever (by time) turned into a footnote, the only game Boston would lose in an otherwise easy World Series triumph.

This didn’t make it a terrible World Series. It just looks worse compared to the previous two, at least by CPS. (WPS says this one was better than the 2016 Cubs-Indians meeting, which frankly leads us to value CPS over WPS.) A good Game Four, in which the Championship Leverage was the highest for the Series, gave a useful boost, while the game with the lowest WPS was minimized by being the one with the lowest Championship Leverage. The series just couldn’t overcome a dominant team performing dominantly.

The March of Time

My postseason retrospectives always have included a look at the length of playoff games, as pace of play has been an enduring concern of mine. It was symbolic that we just saw the longest postseason game in history, because this year we saw the longest average postseason games in history, again.

The problem of long games actually receded during the regular season. Commissioner Manfred’s new limits on mound visits, plus restrictions on between-inning advertising breaks, have produced a drop of nearly four minutes in the average length of a game. This improvement did not carry over into October.

Playoff games lasted an average of 10 minutes longer than the worst previous year, 2014. This wasn’t a matter of one 18-inning slog skewing the numbers, though. (Since 2014 had the 18-inning Giants-Nationals game I mentioned earlier, the two years are even in that regard, and 2018 still finished way ahead.) Measured by inning, this was still the slowest-paced postseason on record, though by just six hundredths of a minute past last year’s mark.

Previously, regular-season and postseason game times have tracked with each other, rising or falling in unison. This postseason broke the pattern and widened a preexisting gap. In 2013, playoff baseball took two and a half minutes per inning longer; in 2015, it was just under two minutes. This year, postseason games took more than four minutes an inning longer than the regular season.

It was the League Championship Series that pushed the pace down to first gear. Innings took a mean of 25.34 minutes in that round, as opposed to 24 minutes flat in the LDS and 23.26 minutes for the Wild Card games. This was perhaps good luck: Had the mostly dull games of the LDS been saddled with the geologic pace of the LCS, viewers would have been repelled in huge numbers. As it stands, carried-over dislike of the LCS’s pace might help to explain the sagging TV ratings of this year’s World Series, though it likely isn’t the sole reason.

So what comes of the slowest postseason ever? Some were speculating in the immediate aftermath of Game Three that Commissioner Manfred—who was at Dodger Stadium for the game, though possibly not all 18 innings—would impose the minor leagues’ new extra-inning rules on the majors in response. I’ve already looked at the effectiveness of those new rules and suggested their success might have opened the door a crack toward implementing them in the majors. A double-length World Series game may have widened that crack.

I find it far likelier, though, that Manfred will implement another pet policy of his. Last year at this time, I said it was all but inevitable he would bring the pitch clock to the majors for 2018. I turned out to be wrong, quite possibly because the players’ union threatened to raise Cain. I’ll chase my losses and predict Manfred will get it in for 2019.

I don’t like this result. It intrudes explicit timekeeping into a game that never has had or needed it before. More gentle methods of quickening pace exist, such as requiring hitters to remain in the batter’s box as was done early in 2015, but that was set aside and seems unlikely to return.

If it’s a choice, though, between the pitch clock and the automatic baserunner in extra innings—not that it necessarily is a choice—I am all for the pitch clock. I’ll take the bad result to avoid the much worse result.

Thus, I end this year’s look back at the postseason on a familiar note: wondering what steps will be taken to make next year’s postseason move a little faster. For the answers, you can do your own calculations or watch this space next year. Your choice.

References and Resources

• FanGraphs for 2018 WPA data
• Craig Edwards, FanGraphs, “Those Disastrous World Series TV Ratings”
• Baseball-Reference for game time and historical WPA data
• Brad Johnson, The Hardball Times Baseball Annual 2014, “The One About the 2013 Postseason”
• Matt Snyder, CBS Sports, “World Series 2018: Could Game 3 be the catalyst for extra innings rule changes?”