Management by Baseball

The Oakland Athletics are at .500 of this writing, and that's a tribute to their pitching and defense because they are 12th of 14 teams in on-base percentage (.308) and dead last with a veritably anemic .228 batting average and a spectre of slugging percentage at .340. By itself, that's not so exceptional as to be worth flapping my lips about it, but some analytical remarks in the San Francisco Chronicle this week brought up an issue I was confronting a client with last month: In the desire to simplify analysis in baseball and beyond, we can miss nuances that are so important that they allow for a confident interpretation of a lie as the truth.

In a complex system, something like a mean average or a mode can be a valuable measure, but only if you think it through and measure context and consequences beyond that number.

Susan Slusser, writing for the San Francisco Chronicle, made or passed on this observation:

In the two games subsequent to Slusser's article, they added two games to that streak, so there hasn't been a single start over the last 20 in which A's starters have given up more than three runs. Pretty extraordinary on first glance -- six of those starts were by ostensible 4th or 5th starters and one by a rookie acting as an emergency fill-in in his first major league start.

We can't really know if this was particularly meaningful unless we look at all the individual components (starters' games), which we will in a few paragraphs after a detour to beyond baseball.

BEYOND BASEBALL
My client was buying some used manufacturing equipment, relatively expensive and with a high transportation charge. He normally would have been decisive and bought something without ever talking with me, but the price scared him a little, and he asked me to talk with him about the comparative virtues and vices of the two- and a half choices he had (the third was really a half -- he was pretty sure he didn't want it, but wanted a third for the comparison). I asked him for the numbers he used and he sent me a text file with four sets of numbers for each unit:  price, presumed remaining life, mean time between failures, and he and I call rated effective throughput [(throughput minus defectives) per hour * % uptime]. 

He then rolled all of these up into an estimated cost for each choice over its life. Smart, as far as he went. Which turned out not to be far enough.

Because he was presuming averages, and the business he's buying this unit to move into is all about turning on a dime with short turnaround orders of custom runs. And in spite of the fact that he knows as well as anyone that mean time between failures doesn't mean you have any idea when the system will fail over time in your shop over time. Some systems break down a lot in their first year on a line and once the weaker parts are replaced are as indestructible and reliable as a DC-3. Some are great unless you read the maintenance protocols closely and realize that you won't be able to keep the unit's pace down to suggested rates. Some just have tolerance that's so fine that your real world system will kill it. And some break down only in clusters -- never for a long time and then with a bunch of loosely cause-'n-effect glitches, go out of service for a long time.

All of these units could have the same MTBF, the same price per unit of rated effective throughput, and have make-or-break differences in how each would fit this particular business. Fortunately, my client had been meeting with other people in this new trade, experienced ones, and his assitant had been researching contract maintenance providers. And what they knew broke his freeze-up. We asked his assistant to find the repair outfits that serviced both of his top choices, and lo and behold, the unit with the slightly less attractive MTBF rating spent less time down per failure because it was almost always the same small group of components that failed and it was routine for repair places to keep them on hand, while the one with the slightly better MTBF unit had to order parts from a warehouse in Mexico and they sometimes got slowed at the border.

In many places, raw MTBF might have delivered the most benefit, but in this particular context, it was less about number of failures as time of repair.

The average is not the reality; it's an artifact... of part... of reality.

AND BACK TO BASEBALL
So how significant was the As' 20-game streak of starters giving up no more than 3 runs? How good were they, and did they show real consistency or were there odd flukes in that run? Let's take a look. To estimate the effectiveness of each start, we're going to use a thumbnail stat left over from early sabermetrics days, Game Score, derived from a formula that includes values for length of start, hits, walks and stirkeouts. Game Score was designed so that an average start would be a 50, better ones getting higher scores and worse ones getting lower. While not "Truth",  it's a useful indicator.

#	DATE	OPP	STARTER	RESULT	IP	H	R	ER	HR	BB	SO	GSc
2	Apr. 2	@SEA	Haren	L 4-0	6	4	4	0	1	1	2	57
3	Apr. 3	@SEA	Blanton	L 8-4	6	5	4	4	1	0	7	53
4	Apr. 4	@SEA	Harden	W 9-0	7	3	0	0	0	2	7	76
5	Apr. 5	@LAA	Gaudin	W 4-3	5	5	2	2	2	0	3	52
6	Apr. 6	@LAA	Kennedy	L 5-2	6	6	2	1	1	2	2	54
7	Apr. 7	@LAA	Haren	L 2-1	7	6	2	1	0	3	2	58
8	Apr. 8	@LAA	Blanton	W 2-1	5.3	5	1	1	0	2	2	54
9	Apr. 9	CWS	Harden	L 4-1	6	5	2	2	2	2	6	58
10	Apr. 10	CWS	Gaudin	W 2-1	5.7	3	1	1	0	3	6	62
11	Apr. 11	CWS	Kennedy	L 6-3	5	5	1	1	0	3	3	53
13	Apr. 13	NYY	Haren	W 5-4	5	4	3	3	0	4	5	48
14	Apr. 14	NYY	Blanton	L 4-3	6.7	5	3	3	1	3	5	54
15	Apr. 15	NYY	Harden	W 5-4	6	5	1	1	0	2	7	63
17	Apr. 17	LAA	Gaudin	W 4-1	7.7	4	1	1	0	2	4	69
18	Apr. 18	LAA	Haren	W 3-0	7	4	0	0	0	0	3	72
20	Apr. 20	@TEX	Blanton	W 16-4	6	7	3	3	0	1	7	52
21	Apr. 21	@TEX	Kennedy	L 7-0	5	3	1	1	0	3	5	59
22	Apr. 22	@TEX	Gaudin	L 4-3	6	4	1	1	0	2	7	65
23	Apr. 23	@BAL	Haren	W 6-5	7	5	1	1	1	1	5	67
24	Apr. 24	@BAL	Braden	W 4-2	6	3	1	1	0	1	6	67
25	Apr. 25	SEA	Blanton	L 2-0	9	6	2	2	2	2	6	71

I use a rule of thumb -- a start with a game score between 46 and 54 I call "medium", the bracketing amounts "better" or "worse". In this run, there was only one start below 50 (a close enough 48). This is 13 "better" and 6 "medium" starts during the three-or-fewer run (Lola) run. These performances weren't fluky. What's fluke-alicious, though, is that the A's could only muster an 11-9 record during this ecstatic time (the last game that continued the streak isn't on this list -- it was finishing up as I was writing).

Are there blemishes here? In the immortal words of Browns fan Curly Howard, "Why Soy-tenlee".  Because while the starters have been legendarily consistent, the bullpen has not been. As abnormal as the starters' consistency has been, it would be almost an order of magnitude more extraordinary for a bullpen to be that consistent...their tolerances tend to be lower (some appearances start with runners on or in scoring position and tougher-than-average hitters up (if they were weak hitters, it would be probabalistically more likely the manager would leave the starter in). And their appearances are shorter, innings fewer, so there's less opportunity to smooth out rough patches into a flat average. The line 1 inning, 0 hits, 1 strikeout, a very lovely reliever line, could occur several times during a start that was crappy. The line 1 inning, 3 hits, 1 walk, 2 runs, an awful relief line, could happen within an overall "better" start. So in short run efforts (in and beyond baseball), what's good or bad has to be considered in context and you have to look at a short stretch with dramatically high or low results as a piece of the whole, not the whole.

In this case, the bullpen's inconsistency undermined the starters' efforts and in part, this was exacerbated by a gaggle of relatively short starts (under six innings). A's relievers gave up 13 runs of their own and 4 runners starters had departed leaving on base during a 23.3 inning stretch over seven games. 

In conclusion on Slusser's newsbyte on the starters, the streak showed actual fine consistency, although in part "three or fewer runs in a start" takes advantage of some pretty short (in some case "minimum" a starter needs to get a win) starts, and a performance like Rich Harden's April 13 start where he gave up three runs in five innings is less impressive than a lot of longer starts where pitchers gave up more. The relative success of this, keeping the A's at .500 even.

SIDE OBSERVATION ON AN A'S INNOVATION
This quick trigger on starters who have not yielded runs or many runners is an interesting approach, not generic. In general, the shorter starts where the pitcher hasn't been knocked out, they have pulled him at below 100, below 90, and even below 80 pitches. Not a common occurrence. Personally, I don't think it's a design for the season, but for early in the season pending Loaiza's return, though I could be wrong, but if I am, I think we'll start seeing lots of churning in their bullpen, with AAA players getting looks in relief roles and probably other acquisitions.

I suspect that having their horse, their most reliable starter Barry Zito, leave through free agency before the season, and then their veteran backup Esteban Loaiza, someone they are counting on for 25-35 starts and 150-225 innings, out at the start of the season with a neck problem, they are trying to win without burning out the remaining starters. And doing a decent job of it, at least on the pitching side, decent enough to be at about the 1/7th mark of the season at .500, one game out of first in a fairly flat division.

On the other hand, we may be seeing the beginnings of an innovation attempt, a new way to handle pitching workloads. If it continues to work as well as it has, we'll see others try to follow, and those others' success will hinge, to a great degree, on how well the context of their talent works in this anti-protocol. Keep watching for those five- to six-inning, 75-90 pitch starts where the starter leaves without having given up much.
END SIDE OBSERVATION

As with my client, more often than not, effective analysis requires more than a look at averages (even intelligently-composed ones). Take a step back from important conclusions long enough to look at stretches of individual data points...there's a lot more useful observation to be had down there.

In the last entry, I spoke about long- and mid-range planning and how baseball is way ahead of other lines of work in finding the balance between and Zen-nothing & the Soviet 5-year plan. In this entry, I'm going to discuss the other mandate for baseball (and you): Real-time plans adjustment.

THE NEW YORK METS: THEY ARE GREAT HERE, TOO
The Mets...all franchises, really, but the Mets especially, are beacons to follow for real-time improvisation.

The Mets got into the National League playoffs having just lost their #1 starter, Pedro Martínez. But they had a great contingency plan in place, because smelling the playoffs early in the season (with a great 28-17 record and a 5 game lead in their division) they had obtained one of the great playoff pitchers in recent history: Orlando Hernández (106 innings, 12-3, ERA of 2.55, and remember this is all against above-average teams). Great advance planning that worked out beautifully.

But their back-up plan was annihilated the day before the playoffs started when Hernández injured himself warming up, and badly enough to be scratched from his start. The Mets had to improvise again, so with tired starters, they picked rookie John Maine. Maine pitched well enough to leave with the lead and they went on to win the game, though cobbling together six pitchers to do it. They followed Maine with Tom Glavine, their surviving ace...an easy enough decision requiring little prestidigitation (three relievers going an inning each), who won the second game. Now, manager Willie Randolph and pitching coach Rick Peterson were cobbling together pieces every day. For the third game, they deployed the erratic Steve Trachsel but for a short time, backed-up by a reclamation project who was having a career yearlet as a middle relief innings-eater, Darren Oliver. Oliver wasn't great but his team came back to finish off the series, this time totaling seven pitchers for the game.

This required knitting together accomplished pitchers, young pitchers, relief role players and a reclamation project. Management spread the work around, and even with the missing pieces, they swept their series and didn't burn out anyone they might have to fall back on in the NLCS. Even entering the NLCS without their Ace or His Replacement, the bullpen would have the benefit of the extra time off (four days) from the sweep.

The Mets pitching for NL Championship Series started with Glavine again (tight game but his 7 shutout innings were key). In game 2, Maine had control issues and in losing, they used six relievers though none enough that they would not be available for the next game (two days later). In game 3 perhaps the most problematic for planning event occurred -- starter Trachsel from whom they needed a better performance, was rocky, took what appeared to be a minor thigh bruise from a batted ball, and pulled himself out of the game with runners on and no one out in the second. Short of starting pitching, Randolph had to call on Oliver who hadn't been particularly effective in the first series -- and Oliver delivered six innings of three-hit ball in a cause that was already lost, enough innings that the Mets didn't need to go to their top three relievers.  They were barely alive, but alive to fight another day and hadn't burned up a pen that might make a difference in a close game they needed to win in the future.

This is important: Oliver's performance and the way the Mets management improvised balanced the low-tolerance mandate  in a short series to "win now", while still preserving resources they needed to have a decent chance to win the next games.

Starter-decimated and down 2-1, the series really could have been over when they sent Oliver "The Culiacan Tomato" Perez to the mound, a once-promising Padre hurler who had fallen apart, for years cruising on fumes and the wishful thinking of the Padres and Pirates before being a late July acquisition the Mets hoped their pitching Merlin, Peterson, could fix. Like Maine, he was effective enough to last enough innings in front of some serious offense that the 3-1/3 innings from three relievers, probably more than ideal, left the pen viable for the next game.

So the series is tied, & the team's best starter, Glavine is up. He has a mediocre start going just 4 innings, Cardinal Jeff Weaver is better, and the Cards go up in games 3-2, only one win to go to the flag. They use the pen, which is effective, carefully & no one gets a ton of work and they have the travel day. In the sixth game, they push Maine out to the mound again, he's great again for his 5-1/3 innings and with the series tied they go back to reclamation project Perez. Perez was great, but the Cards had more offense and won the game and series.

Great improvisation and cool real-time decisions that took into consideration the needs of the future let the Mets get to the last game of the NLCS, an outcome about no-one knowing in advance that would be without Martinez, Hernandez and half a Trachsel, would have imagined possible. The outcome, a halt in their march to the World Series is not the kind that attracts celebration. But Randolph and Peterson played the cards they were dealt optimally at every juncture and stretched what easily could have been a massacre to having the winning run on first base in the 9th inning of the last game -- a little miracle via excellent real-time contingency planning.

BEYOND BASEBALL
In non-baseball organizations, I see contingency planning falling far far short of the Mets' standard. The culture of Baseball is much more evolved, but in your own line of work, you can borrow some of the standard techniques Randolph and Peterson used, and even if you can't execute in real-time to their standard, you can excel relative to any of your peers.

Technique #1 - Always have a range of options for the next step this minute and tomorrow. Just as the Mets blended stars, kids and reclamation projects, you should be prepared for fate to throw you an eephus pitch by having a portfolio of counter-measures. 

Technique #2 - Go against the corporate norm: Count on obvious problems to appear more than unprecedented ones (that is, it's more likely your programmers will design software with a dysfunctional user interface than it is a meteorite twice the size of Prince Fielder will land on your refrigerator unit) and prepare more solutions for the more likely, but at the same time, invest some thinking in what you would do if a less-likely anomaly happened. Mets management probably spent more time worrying about Trachsel having an uneven start than that Orlando Hernandez would break down from stretching the day before the playoffs started. But when Hernandez went down, they had already been thinking about what the most effective responses would be.

Technique #3 - Use people in various roles, apply their efforts before you need to throw them into a high-impact moment. The Mets believed Maine could handle playoff games because they'd stretched him in a number of ways and observed and then measured his responses. In your own workgroups, find out what people know that's not in their job descriptions, try out people on each other's tasks if they're willing and you've been competent enough to carve out some slack to make small investments less risky.

Reminder - One cognitive quirk I see often among even high-capability contingency planners is that a great victory makes them relax their planning cycle. I was subject to this myself as a young manager. I was planning on a project that had staff in all U.S.E.P.A. regional offices except one. I needed to balance workloads between teams, while balancing complex sets of skills and domain knowledge within teams -- a very challenging set of variables, what I call a real "air traffic controller job". We had a crisis in the Dallas office with a sudden audit that a QA vendor decided was urgent (it wasn't) at the same time we got a lot of more-complex-than-usual-work, near the end of an Output-measurement period where our volume was slightly down because we had given extra time off to a person who had had some family issue. We were 1.6 average people off our output rate...a hinky number because we would either fall short of overhire. I knew a few people in Dallas, and asked around if anyone who knew someone with science & stats (or both) in their background who might be available to temp, and as I was talking with one, an old college buddy who sat across a lab table from me in a class or two, I realized she had the almost-perfect skill set. She was working full-time, but agreed to take on a night shift for a healthy rate and profit-sharing. We flew in an associate from another team to simultaneously train and work alongside her, and within a day, we realized we had combined skills and luck for a perfect outcome. All targets met (at the cost of some travel expenses and diluting the profit-sharing pool for the rest of the us). Everyone running to the mound to celebrate. Belief in a perfect (or at least perfectable) world. And then 12 hours later the fecal matter hit the rotary ventilation device in Philadelphia when there was a political battle in the EPA office between some of our client's management, and they said we could no longer work on site (rendering the work totally impossible). And I just froze. For 48 hours, I couldn't think straight. My business partner, when he heard about it, simply called our contract officer at HQ, who fixed it for us. Of course.

The first required complex multivariate planning -- juggling machetes, bowling pins, grenades and meth-smoking weasels. The second only a reasonable phone call to someone with power to fix it. But because it came right after a big "victory", I relaxed to the point I couldn't cope with the next fine-tune required. It was a useful lesson to me, but it's a given beyond baseball for those who don't remind themselves that while some issues are related and cascade, most issues are unrelated -- fixing one has no effect on the other things that can happen. Celebrate, but don't imagine there won't be a next issue right away.

In Baseball, they don't do that...because they can't. Just because you've had the skill to get the best-of-all-possible spare tires (El Duque) doesn't mean it can't take a piece of glass and go flat within a few miles of having to use it. No cause and effect between one issue and (the inevitable) next one. Just plan, adjust, implement, adjust, plan.

Another reminder - Occam's Razor holds here. The fewer factors you can have affecting choices, the fewer possible things there are to require addressing. In the EPA case I described, we had teams, skills, EPA HQ, EPA regional management, and the work-audit vendor. If it had just been the first three, the potential for issues would have been roughly one-third the total of the five together. We didn't have much choice either in how many groups we dealt with. But in planning and executing projects in general, simplification, slack, and reduction or delay of what I call "hood ornaments" (nice-to-haves that could be solid additions later) give you an environment where fewer issues are around to arise.

Outside baseball, one of the most common managerial freeze-ups occurs over planning initiatives - whether they are quick mini-projects or big enterprise-wide endeavors or mega-sized military operations. Almost all organizations have a terrible time finding the viable middle between the Scylla of rigid Soviet Five-Year Plans (based on wishful thinking or lack of imagination or, scarily, both) and the Charybdis of ad-hockery, a refusal to apply any planning ("because we can't be assured any outcome that happens will be one we could have planned for").

Baseball kicks other industries' collective axe. Baseball's default is mastery of rigorous, pre-meditated planning out of the future, concurrently with a lot of on-going contingency planning, combined with real-time improvisation at the time of execution. It's a given...if managers can't do this, they won't get a role like that in the majors or if they do, hold on to it very long.

ONE STERLING EXAMPLE: THE NEW YORK METS
The Mets this weekend provided a perfect illustration of the first of the three planning efforts I just mentioned. They had a final Spring Training game on Saturday and the season opener on Sunday. Optimization says current ace Tom Glavine needs to start the season opener on Sunday, but who gets the start in the Saturday game?

The easy answer is "some throwaway".  The easy answer here is the wrong answer. True, you don't want to use any of your core starters who will be pitching in games that count in the following week(s). A game that doesn't count, against the lowly Tampa Bay Devil Rays suggests at first a minor-league pitcher, because who cares? So many corporate decisions are made in that spirit, which is why so many corporate decisions are feeble.

Baseball doesn't do that. In Baseball, every decision has been planned for in advance, every possible side-benefit, foreseeable contingency, and cost considered. The answer(s) will result from careful consideration of all those factors.

So the answer Saturday was "Mike Pelfrey", neither a throwaway nor a core starter, but a rookie pitcher Mets management needs/wants to get more work against major leaguers. They hope both that he will prove to be a long-term quality starter, and that when they need a spare tire for the rotation during the season, he will be up to the task. He needs innings against major leaguers for both purposes. Baseball has figured out (why hasn't the corporate world?) that it's easier to master a skill set starting with lower-pressure situations and then ratchet up the difficulty to ascertain the subject's current ceiling and figure out what the next training required will be.

So Pelfrey got four starts last year, and against "easy" opponents, Florida, Houston and Cincy, all below average in runs scored per game. He got four starts in Spring Training this year,  (and how's this for a fascinating but meaningless coincidence: he had an identical ERA of 5.48 in his small 2006 major league line and in his 2007 Spring), and induced about twice as many groundballs as flyballs both last season and in this Spring. He'd had a good Spring until Saturday. He could have a good game and cement the possibility of being the team's "5th starter" or his work would show what he needed to work on.

There was another important factor, actually a blessing, in the decision. The Mets upcoming schedule. 

The late 1950s Orioles under Paul Richards publicly acknowledged they planned out in March every starter for the entire season. The practice probably goes back farther, or at least for a big hunk of the season. Richards knew darned well that the odds of that holding, coming to be exactly as charted out, was infinitesimal, but the long-range planning exercise was indispensably useful later on. The almost-certainties (days off in the schedule, opponents on given days) allow for an exercise where you optimize using your top starters the most, line up individuals for match-ups against specific teams where useful and possible, and spread out rest. Having already marinated one's self in the details, when something needs to be changed later, there's already a certain level of intimacy with the fine points, and a preparation for the nasty stretches that would make contingency responses challenging.

Every day, team management (in the Mets' case, Willie Randolph, Jerry Manuel, Rick Peterson & Omar Minaya) will spend a at least little time thinking about what to do tomorrow if the feasible disruptions or disasters happen today. The existence of the pre-season year's pitching chart doesn't lessen the practice -- it just makes it more supple and more quickly adjustable.

Here's what key data points on the April part of the Mets' chart might have (and probably) looked like.

The days off allow both for pitchers to generally throw on four days rest (a generally useful interval) and also mean that the presumed-best four starters can handle all the work through Monday the 16th without having to work on short rest. The two days off in the first week mean the Mets' #1, Glavine, could start with 4 days rest on the 6th, but I suspect not -- the game is in Atlanta, and he has not performed as well against his old team as the rest of the league, and less-well in Atlanta than when he faces the Braves in New York. And it's a long season. As Cardinals' pitching coach Dave Duncan said in a New York Times piece yesterday, “You do all of the things that you can to give <the pitcher> the best chance to be at his best during the course of the season. Part of that is to not run him into the ground at any particular point in time.” It's a long season, and squeezing out an extra start for a 41 year old pitcher in April can have negative consequences in August or September. It's a marathon, not a sprint. But Glavine could start that game on the 6th, and might...especially if...

BATS IN THE PELFREY
Pelfrey, as the #5 starter, didn't need to be on the big team's roster to start the season. As you can see, injury aside, the Mets don't look to need a 5th starter until the 13th game of the season, on the 16th. In the meantime, he can be at AAA, getting more work based on what he needs the most. And Saturday, he got Fallujah-ed by the D-Rays. He has no shortage of immediately-remembered things on which to get help.

The Mets management, like any major league team's, will observe Pelfrey's progress over the next couple of weeks. When the time comes to choose a #5 starter in mid-April, it will probably be Pelfrey. But they might riff off that default setting. If he struggles, or if someone who has started effectively in the majors before (Aaron Sele now in the big club' bullpen, Chan Ho Park down in AAA) meets management's requirements and proves ready, we might not see Pelfrey. In Baseball, plans are formed AND fluid.

Beyond Baseball, there's a lot to learn from these practices. A completely-elaborated long-term plan doesn't preclude creative adaptation -- in fact, when properly formed and followed, it enhances the potential for creativity, as my esteemed colleague Diana Wynne has argued. It easier to act quickly and improvise sensibly when a team has already steeped itself in the full details of the full schedule -- they already have confronted gotchas and knowable barriers.

Baseball has a ton to teach non-baseball organizations about long- and mid-term planning. But in addition, Baseball, and the Mets specifically, have a ton to teach about real-time improvisation, lessons I'll point out in the next entry.