<$BlogRSDUrl$>

Monday, January 25, 2010

Deming's Disciple Unleashes Brewers' Fermentation  

“It does not happen all at once. There is no instant pudding.” -- W. Edwards Deming

In general, North American businesses are incapable of taking advantage of North America's #1 ever native management consultant. W. Edwards Deming revolutionized a variety of industries, primarily manufacturing, around the world by promoting kaizen, continuous improvement through mostly small changes.

Kaizen doesn't fit well into the structure of the large publicly-owned corporation as practiced most places, especially in the U.S., in Russia or in Red China. That's for two main reasons.

One, the approach generally puts workers into a leading position in change design, and that doesn't fit well in the normally very top-down corporate designs of corporations in those three countries...allowing that to happen would disrupt management to line staff compensation ratios.

Two, continuous improvement relies mostly on small, quickly deployed changes and there's little career-building credit to accumulate to ambitious executives for any single one. So mega-projects and grandiose programs garner more attention and enthusiasm, and eventually compensation rewards.

Because of those dysfunctional barriers, Deming can be hard to implement in North American corporations, but he's even harder to read -- his materials are a barrier to success.

One of Deming's most successful U.S. disciples is the Kaizen-pushing Rick Peterson, the Milwaukee Brewers' new pitching coach. And one of the reasons he's so successful, even managing process and people in an industry where the average education is much lower than average, is because he knows how to use accessible ideas to describe the very lofty things he's achieving.

On the surface, it looks like he has a long way to go. The Brewers were an N.L. worst in starting pitcher ERA and some other indicative stats last year.

Tm R/G CG SHO QS% GmScA ? RS/GS
SFG 3.77 11 3 52% 55 4.1
LAD 3.77 1 1 51% 54 4.8
ATL 3.96 3 0 61% 54 4.5
STL 3.95 8 4 54% 53 4.6
CHC 4.17 3 2 58% 53 4.4
ARI 4.83 4 1 55% 51 4.4
COL 4.41 5 2 58% 51 5.1
PHI 4.38 8 5 51% 51 5.1
LgAvg 4.49 5 2 50% 50 4.5
SDP 4.75 2 1 48% 49 3.9
FLA 4.73 5 1 46% 49 4.8
CIN 4.46 6 4 49% 49 4.1
NYM 4.67 3 2 48% 48 4.2
PIT 4.77 5 2 52% 48 4
HOU 4.75 5 1 44% 48 4
WSN 5.4 6 1 39% 46 4.4
MIL 5.05 1 0 40% 45 4.9
Baseball-reference.com

According to the Milwaukee Journal Sentinel, Peterson recently conducted an organization-wide pitching symposium in Arizona, and the reporter wrote about the core changes the team was making. Not a mega-project, not a self-aggrandizing grandiose Shlock 'n Awe publicity-fest. Kaizen.

Non-extreme makeover

Many residents of Brewer Nation are fighting panic attacks regarding the team's starting rotation.

After the Brewers finished last in that department in the National League with a 5.37 earned run average in 2009, general manager Doug Melvin replaced just one starter, punting Braden Looper and signing free agent Randy Wolf.

There is still time to add another starter, and Melvin stays in touch with the agents of certain pitchers, such as Doug Davis. But, should the rotation include just the one change, new pitching coach Rick Peterson is determined to make it work.

"You take a look at each guy and say, 'What are incremental differences we can make to help him win?' " said Peterson, who detailed his plan during the organization's recent pitching symposium in Phoenix.

"It's like compound interest. It doesn't grow a lot in one day. It grows over time. If you can get guys to make incremental improvements, you get significant improvement over time.

Simply put, it's classic Deming, but because Rick uses an apt & broadly understood analogy, and because he's boiled the explanation down to just the simplest apparent feature, it's easier to "get", ergo more likely to be internalised in context, ergo more likely to be implemented properly, ergo more likely to trigger significant improvements.

But just how much value would the tiny incremental improvement Peterson posits actually mean in winning potential?

The Kegmen were 80-82 last year. They scored 785 runs (a swell 3rd best of 16 teams) while allowing 818 runs (a scary-bad next to worst). How far away were Milwaukee's last-in-the-league-for-ERA starters from allowing an average number of runs?

If Rick's increment were, for example, 1 fewer hit per series, 1 fewer walk per series, 1 additional strikeout per series, it seems on the surface not an unreasonable target. The average 2009 team yielded 27.4 hits per series, so it's not full-tilt Nam-myoho-renge-kyo to believe better pitching could trim that to 26.4 per series. And aiming to trim the average number of walks per series from around 10 to about 9 seem more ambitious, but the pitcher has a little more control over this, so actual training and focus might yield that.

But how much difference would that 1 hit less, 1 walk less and 1 strikeout more per series (1-1-1) make? Glance at the table below in which the conceptual/incrementally-tweaked Brewers are yclept MIL10.

Tm IP H HR BB IBB UIBB SO WHIP H/9 HR/9 BB/9 SO/9 SO/BB
WSN 1424 1533 173 629 59 570 911 1.52 9.7 1.1 4.0 5.8 1.5
HOU 1430 1521 176 546 56 490 1144 1.45 9.6 1.1 3.4 7.2 2.1
MIL 1435 1498 207 607 60 547 1104 1.47 9.4 1.3 3.8 6.9 1.8
PIT 1418 1491 152 563 37 526 919 1.45 9.5 1.0 3.6 5.8 1.6
PHI 1455 1479 189 489 31 458 1153 1.35 9.1 1.2 3.0 7.1 2.4
ARI 1447 1470 168 525 27 498 1158 1.38 9.1 1.0 3.3 7.2 2.2
NYM 1426 1452 158 616 60 556 1031 1.45 9.2 1.0 3.9 6.5 1.7
COL 1438 1427 141 528 51 477 1154 1.36 8.9 0.9 3.3 7.2 2.2
FLA 1446 1425 160 601 60 541 1248 1.40 8.9 1.0 3.7 7.8 2.1
SDP 1450 1422 167 603 58 545 1187 1.40 8.8 1.0 3.7 7.4 2.0
CIN 1458 1420 188 577 36 541 1069 1.37 8.8 1.2 3.6 6.6 1.9
STL 1440 1407 123 460 23 437 1049 1.30 8.8 0.8 2.9 6.6 2.3
ATL 1462 1399 119 530 59 471 1232 1.32 8.6 0.7 3.3 7.6 2.3
CHC 1445 1329 160 586 46 540 1272 1.33 8.3 1.0 3.6 7.9 2.2
SFG 1446 1268 140 584 49 535 1302 1.28 7.9 0.9 3.6 8.1 2.2
LAD 1473 1265 127 584 68 516 1272 1.26 7.7 0.8 3.6 7.8 2.2
LgAvg 1444 1425 159 564 49 515 1138 1.38 8.9 1.0 3.5 7.1 2.0
MIL10 1435 1446 207 555 60 495 1156 1.39 9.1 1.3 3.5 7.3 2.1

Not too shabby. IF Peterson's increment was the modest 1-1-1 for the staff, the Brewers would have been roughly composite league average, and a league average pitching staff yielded 727 runs (not the 818 the real 2009 Brewers did). If the 2009 Brew Crew scored the 785 runs they actually did, while yielding a league composite average 727 runs, then according to a thumbnail Pythagorean estimate, one might expect them to win 87 games, a seven-game improvement. It's not, in and of itself, enough to take over the world, but it's Deming-like...an inexpensive, targeted low-overhead quality improvement that frees up slack for attention to other things.

What looked like a ferocious challenge (getting the Brewers' awful starting pitching performance to average) yields to small, incremental changes plotted studiously and executed relentlessly. Deming

But let's get even more specific for a moment. 2009 NL starting pitchers averaged 5.8 innings per start. The 2009 Brewer starters gave up 566 runs in 891 innings (3.68 runs/5.8 innings of starting pitching). Had they been merely adequate enough to yield the league composite average for NL starters, it would have been 2.84 runs/5.8 innings of starting pitching. Over 162 games, league average yields 136 fewer runs than the actual 2009 Brewer starters.

Small, incremental changes executed daily yield to serious, competitive returns. And classic Deming in that the small quality improvement might yield other advantages as well. In this 2010 Brewer case, there's an argument to be made that if the starting pitching is "better", it will put less demand on bullpen innings, meaning incremental available innings can be shifted to "better" relievers, and that could lead to better results. More classic Deming.

Deming's disciple is Peterson, and Peterson is lucky to be working for as experimental an executive as Brewer GM Doug Melvin, but even the non-Melvin shops in Baseball are far more capable of following and successfully implementing Deming than most corporations or academic or non-profit organizations are.

If you're not applying Deming in this economy, then what the heck are you waiting for?


This page is powered by Blogger. Isn't yours?


free website counter