DIAMOND MIND NEWS
August 30, 2000
Written by Tom Tippett

Welcome to the fourth edition of the Diamond Mind email newsletter for the 
year 2000. Through these newsletters, we will try to keep you up to date 
on the latest product and technical information about the Diamond Mind 
Baseball game, related player disks, and our ongoing baseball research 
efforts. Back issues are available on our web site (www.diamond-mind.com).

If you don't wish to receive these messages in the future, please send an 
email response with the subject line "unsubscribe". We'll immediately 
remove your email address from the list. And if you know someone who would 
like to subscribe to this newsletter, we'll be happy to add them to the 
mailing list if they send us an email message with the subject line 
"subscribe" and their name and street address in the body of the message.

Topics for this issue:

Clutch hitting
Version 8 update


Clutch hitting
==============

Last December, Rob Neyer of ESPN.com wrote an article that sums up the 
attempts that have been made to demonstrate which players can properly be 
regarded as clutch hitters. That article is no longer available on 
ESPN.com, but Rob was gracious enough to grant us permission to publish it 
on our web site, and you can find it at:

http://www.diamond-mind.com/articles/neyerclutch.htm

Regular readers of this newsletter know that I highly recommend Rob's 
column, which can be found at the following address:

http://espn.go.com/mlb/columns/neyer_rob/42798.html

Rob presents both sides of the debate about whether clutch hitting exists, 
so I won't get into the details here. But it boils down to two points of 
view. The media takes for granted that certain players can be counted on 
to rise to the occasion and that you can identify those players by looking 
at batting averages in the late innings of close games or with runners in 
scoring position. On the other hand, the baseball analysis community 
hasn't been able to find any evidence to suggest that this is true.

This is relevant for Diamond Mind Baseball because we include clutch 
ratings for hitters and jam ratings for pitchers on our season 
disks. Because there was no evidence that clutch hitting really exists, my 
original design did not include clutch and jam ratings. But when I signed 
the deal that led to the game being marketed by Pursue the Pennant from 
1987-94, the folks at PTP insisted that I add these ratings.

The best argument for including them goes like this. If a real-life team 
happened to have three or four guys who compiled better stats in clutch 
situations (however you define them) than in non-clutch situations, chances 
are they won more games than normal given the talent on their roster. So 
if we take the side of the researchers who say that clutch hitting doesn't 
exist, we'd leave these ratings out of our game, and teams like this could 
fall short of its real-life win total in season replays.

But just because someone batted thirty points higher in 'clutch' situations 
than in other situations, it doesn't necessarily mean that he was a 
prime-time player. Most regulars get only 50-75 atbats in clutch 
situations in a season. With any group of atbats of this size, you'll have 
no trouble finding players who were up and others who were down solely due 
to chance. So how do we tell the difference between a player who got lucky 
and a player who rose to the occasion?

And what are clutch situations anyway? If you define them as the late 
innings of close games, then it's not a clutch effort when a hitter blows 
open a close game with a three-run homer in the sixth. If you define them 
as any situation with runners in scoring position, then it's not a clutch 
effort when a pesky leadoff hitter draws a walk and goes on to score the 
tying run. I submit that there are a lot more clutch situations than the 
media tends to include in their 'analysis'.

Even if we could come up with a consensus on how to define clutch 
situations, and even if there was some compelling evidence that certain 
players consistently come through in these situations, I'm still left with 
a troubling question. If someone compiles better stats in clutch 
situations, doesn't that mean he's not performing at his best in other 
situations? Doesn't that suggest that he's coasting or failing to focus 
adequately in the early innings or when nobody is on base?

(By the way, most of the research has focused on attempts to find players 
whose stats improve in clutch situations, but it's possible that clutch 
performers distinguish themselves by maintaining their levels while others 
decline. In the past three years, batting averages have been 6-14 points 
lower in the late innings of close games than in all situations 
combined. This may reflect nothing more than the fact that you're 
generally facing the opposition's best pitchers in these situations, but 
there might be some clutch-related stuff involved, too.)

I could go on, but suffice it to say that I'm not convinced that clutch 
ratings belong in a game like ours that is designed to reflect what really 
happens in baseball. But I can't honestly say that this matter has been 
fully researched either, and until we have the time to do a comprehensive 
study of our own, I have to admit that it's possible that clutch performers 
do exist even though the baseball research community cannot prove 
it. Besides, these ratings have been in the game for thirteen years, and 
they're going to stay.

Maybe this matters only to me, but I've chosen to 'solve' my dilemma by 
making the clutch system optional in version 8. When setting up your 
league rules, you can choose to use the clutch and jam ratings or ignore 
them. That way, if you think I'm nuts to dismiss the notion that clutch 
hitters exist, you can continue to use them. If you are among the 
skeptics, you can turn them off.

If you use the clutch ratings, you probably want to know what effect they 
have, so if you haven't already heard this, I'll repeat it here. They come 
into play in the late innings of close games regardless of whether there 
are runners on base or not. They do not have a large effect on the outcome 
of the batter-pitcher confrontation, however, and I would never choose to 
use a weaker player over a better one just because he has a superior clutch 
rating. That said, you will gain a small advantage if you have a clutch or 
jam rating in your favor.


Version 8 update
================

I'm happy to report that we began field testing version 8 about three weeks 
ago. In that time, we've been focusing much of our effort on testing the 
game and fixing the bugs that have been reported, but we've also found time 
to make good progress on the last few features that we're adding.

The one I'd like to talk about now is the ability to save a game in 
progress and resume it later. During a recent visit with my wife's family, 
my 11-year-old nephew spent much of the time playing with version 8 on my 
notebook computer. On several occasions, he was in the middle of a game 
when we all needed to leave for a family outing. Because it tends to 
overheat, I prefer to shut this computer off when I'm going to be away for 
a while. But turning it off meant telling my nephew that he'd have to 
forget about the game that was underway.

It had always been part of the plan to add the ability to save a game and 
resume it later, and this experience reinforced the need for this 
feature. So, in version 8, we've added a menu command that allows you to 
save any exhibition or league game. Because league games involve 
year-to-date stats, injuries, and fatigue information, you won't be allowed 
to resume that game if you have subsequently changed the rosters of the two 
teams, played or imported any other games involving the two teams, or done 
anything else that could cause the information in our database to get out 
of synch. A saved exhibition game stands alone, so it can always be resumed.

While we were adding this capability, we also added a feature in which the 
state of the game is automatically saved after every play. The purpose is 
to provide a safety net for anyone who suffers an untimely problem with 
their computer. If Windows crashes or is shut down during a game, or if 
you're playing a game online with the help of software like NetMeeting and 
your connection is dropped, you'll be able to resume the game right where 
you left off.

The version 8 to-do list is getting shorter every day, and you'll be the 
first to know when we have a formal ship date to announce. In the 
meantime, we're going to finish up the handful of new features that are 
still in the works, and we'll tell you about them in the next newsletter.

Tom