Examining the A’s Epic Collapse

One of the biggest stories of the season’s second half has been the historic decline of the Oakland Athletics. They are flirting with accomplishing the extremely difficult feat of having the best record in baseball at the All Star break, and then missing the playoffs. Winning the final two games of their pivotal series with the Seattle Mariners this past weekend has sharply decreased the likelihood of that worst case scenario, but the collapse has been stark nonetheless. It’s convenient to tie the A’s second half results to the departure of Yoenis Cespedes in the Jon Lester trade, but the reality is a bit more complicated than that. There are many factors in play, but arguably the foremost among them has been the precipitous fall of two of their key offensive players – Derek Norris and Brandon Moss.

As we stand here a week and a half away from the end of the regular season, one can still make a very strong case that the A’s are the best club in the major leagues, despite the fact that they currently trail the Angels by 10 1/2 games, through Tuesday’s night’s games. Their run differential remains the game’s best, and over a large sample size, they have proven to be a balanced, resilient club. They are 3rd in the AL in runs per game, and 2nd in runs allowed. They play strong defense. They don’t have wasted spots on their roster, platooning often in the lineup, and getting valuable contributions from both the back end of the rotation and bullpen.

The A’s record crested at 28 games over .500, at 72-44, on August 9, though their winning percentage peaked at .625 a couple weeks earlier, on July 27. They are 18-28 since the former date, and 11-23 since the latter. Obviously, this is right around the time of the Cespedes trade, on July 31. Usually, when a contender acquires a high-end rental player like Lester at the deadline, minor league pieces are exchanged in return.

By dealing Cespedes, the A’s did weaken their lineup to strengthen their rotation. They did receive Jonny Gomes in the deal, but unsurprisingly, he appears to be toast, batting .250-.317-.269 since the deal. Sam Fuld was also re-acquired that same day, and has had to play quite a bit, batting a puny .217-.272-.344 in two stints with the A’s this season. Certainly, there has been an offensive dropoff with the departure of Cespedes, but there is that slight mitigating effect – sarcasm intended – of the acquisition of Lester and his 2.30 ERA in 62 2/3 innings since the trade. The offense was hurt, but the team significantly helped by the deal. As a club, the A’s were batting .254-.329-.406 through July 29 – since that day, they have hit only .223-.295-.335. It has been run creation, not prevention, that has keyed the club’s decline.

One under-the-radar reason for the A’s offensive decline has been the absence of John Jaso, placed on the DL on August 24 with a concussion. From July 29 until that date, Jaso went 8 for 52, batting .154-.214-.231 over that span. Jaso is no superstar, but he is exactly the type of versatile, undervalued piece that the A’s productive offense has been built upon. The timing of his decline exactly mirrors that of the club, but every contender has injuries, and many of them have occurred to players more valuable than Jaso. Another mitigating factor here has been increased playing time for Steven Vogt, who has contributed more than the A’s ever could have asked.

This brings us back to Norris and Moss. Both players were acquired in two of the more common ways utilized by the A’s. Gio Gonzalez was becoming too expensive for the A’s taste, so they dealt him to the Nationals for a substantial package that included Norris. Moss was a humble minor league free agent signing prior to the 2012 season, an undervalued asset that the A’s identified as a potential breakout power performer.

On August 2, Norris woke up with a .303-.402-.481 line, and Moss had a line of .266-.347-.526 as recently as July 24. Obviously, these guys were a big part of their great start, and they have been an equally large part of their recent struggles. Let’s first examine their overall 2014 plate appearance outcome frequency and production by BIP type data to get a better feel for their true talent level. First, the frequency data:

FREQ – 2014
Norris % REL PCT
K 19.7% 97 54
BB 11.8% 153 89
POP 9.4% 122 69
FLY 27.3% 98 42
LD 18.0% 86 10
GB 45.3% 104 69
————– ———- ———- ———-
Moss % REL PCT
K 26.5% 131 83
BB 11.0% 143 84
POP 11.0% 143 81
FLY 37.8% 135 96
LD 20.7% 100 48
GB 30.4% 70 3

There’s actually a lot to like in Norris’ full-season frequency profile. HIs BB rate is quite high, with an 89 percentile rank, providing some margin for error with regard to batted-ball authority. Though his popup rate remains quite high (69 percentile rank), it is way down from his massive 17.2% popup rate in limited 2013 duty. His liner rate (10 percentile rank) is low, but liner rates tend to fluctuate more than those of other batted-ball types, so positive regression could be in his future.

Moss is what he is, an extreme fly ball hitter (96 percentile rank). In my Chris Davis piece earlier this week, it was noted that Davis and Moss, along with Colby Rasmus, were the three most extreme fly ball hitters in the game in 2013. Such hitters are extremely risky, and the vast majority of the 17 2013 MLB regulars who had more fly balls (excluding popups) than grounders have struggled quite a bit in 2014. Moss strikes out a ton (84 percentile rank), but like many power hitters gets his walks (83 percentile rank). With such a high K rate, it is imperative that Moss make thunderous contact when he does put the ball in play.

To assess the relative authority of both players’ contact, let’s take a look at their overall 2014 production by batted-ball type, before and after adjustment for context:

PROD – 2014
FLY 0.299 0.687 98 66
LD 0.523 0.841 75 113
GB 0.342 0.378 202 126
ALL BIP 0.339 0.518 109 94
ALL PA 0.262 0.350 0.401 117 104
—————- ———- ———– ———- ———- ———-
FLY 0.363 1.062 194 246
LD 0.726 0.919 116 115
GB 0.242 0.264 100 110
ALL BIP 0.345 0.637 136 153
ALL PA 0.239 0.323 0.442 116 128

Both players’ actual production on each BIP type is indicated in the AVG and SLG columns, and is converted to run values and compared to MLB average in the REL PRD column. That figure then is adjusted for context, such as home park, luck, etc., in the ADJ PRD column. For the purposes of this exercise, SH and SF are included as outs and HBP are excluded from the OBP calculation.

There are major disconnects between Norris’ actual and adjusted production for each batted ball type. He’s been quite lucky on fly balls (98 REL PRD vs. 66 ADJ PRD), a fact that has been largely offset by his bad fortune on liners (75 REL PRD vs. 113 ADJ PRD). What cannot be ignored, however, is the extreme good luck he has experienced on ground balls. Norris is batting a stunning .342 AVG-.378 SLG on grounders for the season – a 202 REL PRD figure, over twice as productive as the average MLB hitter. This, from a relatively, slow, righthanded-hitting catcher. He does hit his grounders (and liners) fairly hard – adjustment for context only brings his ADJ PRD on grounders down to 126. Overall, his REL PRD on all BIP is above league average at 109, but is adjusted down to 94 for context, largely due to the grounder disparity. Adding back the K’s and BB’s is a net plus for Norris, raising his ADJ PRD to 104 – above league average production at a defense-first position is a very good thing.

Moss hurts the baseball, as a high-K guy must. He has a strong 194 ADJ PRD on fly balls, and such performance in a relatively pitcher-friendly park elevates it even further to a 246 REL PRD. Like Norris, Moss hits his liners and grounders quite hard, with ADJ PRD figures of 115 and 110, respectively. Overall, his above average fly ball frequency and authority drives his REL PRD on all BIP to a very high 153. His fairly extreme K rate takes a toll when the K’s and BB’s are added back, dropping his overall ADJ PRD to 128.

The full-season numbers paint both Norris and Moss as the productive offensive performers they are, both overall and relative to their respective positions. Let’s split both players’ frequency and production data into 1st and 2nd half data, using August 2 as Norris’ cutoff, and July 24 as Moss’, to see what has gone wrong.

FREQ – 2014
Norris – 1st % REL PCT
K 16.9% 83 42
BB 14.3% 186 98
POP 9.0% 117 66
FLY 27.7% 99 45
LD 20.9% 100 48
GB 42.9% 98 59
————– ———- ———- ———-
Norris – 2nd % REL PCT
K 25.5% 126 79
BB 7.1% 92 35
POP 10.3% 134 76
FLY 26.5% 95 35
LD 11.8% 57 1
GB 51.5% 118 90
————– ———- ———- ———-
Moss – 1st % REL PCT
K 24.0% 118 73
BB 9.4% 123 70
POP 10.9% 141 80
FLY 37.9% 136 96
LD 20.2% 97 37
GB 31.0% 71 4
————– ———- ———- ———-
Moss – 2nd % REL PCT
K 32.9% 162 98
BB 14.2% 184 98
POP 11.8% 153 84
FLY 37.3% 134 95
LD 23.5% 113 82
GB 27.5% 63 2

Some really interesting stuff here. Norris’ big problem has been the utter breakdown in his K and BB rates, going from 16.9/14.3 ratio in the 1st half to 25.5/7.1 in the 2nd. There have been some variations in his other frequencies, primarily, the absolute crashing of his liner rate to 11.8% after August 2, but the K/BB breakdown is the key item.

The biggest piece of info in Moss’ frequency profile is the explosion in his K rate, from 24.0% in the 1st half to 32.9% in the 2nd. This puts extreme pressure upon his batted ball authority for him to have any degree of success. The BIP production by half data should give further insight as to why both hitters have struggled so much of late:

PROD – 2014
FLY 0.327 0.816 129 90
LD 0.556 0.889 84 114
GB 0.382 0.421 251 120
ALL BIP 0.364 0.576 129 100
ALL PA 0.291 0.394 0.461 151 122
—————- ———- ———– ———- ———- ———-
FLY 0.222 0.333 35 20
LD 0.375 0.625 40 106
GB 0.257 0.286 115 139
ALL BIP 0.292 0.406 74 84
ALL PA 0.212 0.268 0.295 66 74
—————- ———- ———– ———- ———- ———-
FLY 0.393 1.191 237 284
LD 0.680 0.880 104 109
GB 0.247 0.273 105 117
ALL BIP 0.364 0.715 161 180
ALL PA 0.265 0.336 0.522 141 155
—————- ———- ———– ———- ———- ———-
FLY 0.211 0.421 42 97
LD 0.917 1.083 175 143
GB 0.214 0.214 73 76
ALL BIP 0.280 0.373 65 76
ALL PA 0.167 0.291 0.222 61 68

The stark breakdown in Norris’ fly ball authority is the most eyecatching aspect of his profile. He is batting just .222 AVG-.333 SLG on fly balls in the 2nd half, an amazing 35 REL PRD that is adjusted even lower for context to 20. Norris is completely unable to elevate the ball in the air for distance up the middle and to the opposite field. He has hit a grand total of two fly balls at 90 MPH or higher in those directions all season, and of late, Norris has had a much harder time pulling the ball in the air. Through August 2, Norris hit 28 balls in the air to LF-LCF, and 14 to RCF-RF. Since then, he has hit 7 to LF-LCF and 7 to RCF-RF. Norris’ body and bat are slowing as the season progresses.

Knowing that he is unable to hurt them to fully two-thirds of the field, pitchers are learning that they can throw strikes to the outer half vs. Norris and not be hurt. His BB rate has plunged, and coupled with his almost nonexistent liner rate and the expected regression of his performance on grounders – .342 AVG-.378 SLG in the 1st half, .257 AVG-.286 SLG in the 2nd – voila, you have Norris’ 2nd half.

As for Moss, as it is for Chris Davis, it’s all about the pulling. Moss is arguably the most extreme pull hitter in the game today, and that inherently makes him streaky and risky. Looking at Moss’ 1st and 2nd halves is like looking at Raul Ibanez‘ 2013 and 2014 seasons. Not coincidentally, Ibanez was also on the list of extreme fly ball hitters in 2013. Moss has totally lost his swing in the 2nd half, a swing that is grooved to drive mistake fastballs on the inner half over the RF fence, and little else.

He’s gone from .393 AVG-1.191 SLG on fly balls in the 1st half to .211 AVG-.421 SLG in the 2nd, though context and a very small sample size bumps his 2nd half fly ball ADJ PRD figure up to 99. Moss needs to pull in the air to excel, and he has struggled to do so in the 2nd half – 72% of his fly balls were hit to RCF-RF in the 1st half, but only 50% have been pulled in the 2nd half. Moss also hits extremely “high” fly balls, another risky feature covered in the Davis article. If anything, Moss has been incredibly fortunate on ground balls, as he is the most obvious overshift candidate out there, with a 13.17 pull ratio (79 grounders to RCF-RF, 6 to LF-LCF). If he starts to hit .120 on grounders like Davis, Moss could disappear in an instant.

The A’s have largely built their club around players such as Norris and Moss, and deserve a great deal of credit for doing so. Both possess clear strengths, but also have clear shortcomings. As their careers have progressed, pitchers have begun to pound their weaknesses, and the onus is now on them to make the necessary adjustments. Moss has homered in two of the last three games, and even 70% of the first half version of Norris is valuable because of the dearth of offense at the catching position, so all is far from lost. We should realize, however, that perhaps the A’s and their offense weren’t quite as good as they appeared to be a couple months ago, and that their full season numbers only now represent their true talent level.

Newest Most Voted
Inline Feedbacks
View all comments
Paul M
9 years ago

it gets worse. Coco Crisp got a nagging neck injury in early July– he hasn’t really been healthy since, and he fell off the table. Craig Gentry also has been injured and missed significant time. More recently Vogt has been hurt, though his drop in production the past 6 weeks or so has been more of a classic regression to mean. The combination of Sogard, Punto, Callaspo (seeing him stride to the plate on a 4-40 tear with bases loaded and one out in 8th inning last night was simply criminal by Melvin– with Moss available!!) Gentry, Fuld, Gomes, Soto is simply one 230-300-350ish hitter after another– and for several the production hasn’t even been that good. Lowrie has also slumped bigtime (he was the other culprit with bases loaded in 8th last night). Basically Donaldson and Reddick are the only two A’s regulars– out of a total pool of 14-15, given platoon and injuries– that have kept their production level since mid July… And no one has improved these past two months…

9 years ago
Reply to  Paul M

Two words – sample size. 4 for 40 isn’t really meaningful unless you can identify a reason for it.

Philip Christy
9 years ago
Reply to  Doug

How about a sample size of 442 PAs? .228/.296/.296, 71 wRC+.

Buns Slugsworth
9 years ago
Reply to  Philip Christy

Jeter’s line is .249/.298/.297 in 595 plate appearances.

9 years ago
Reply to  Philip Christy

So Callaspo should retire too?

9 years ago
Reply to  Philip Christy

He should have retired 4 months ago. That way Melvin wouldn’t pinch hit him with the bases loaded and one out in a 1-run game.