|
Diamond Mind Email Newsletter #10

October 25, 2000
Written by Tom Tippett
Welcome to the fifth 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.
Topics for this issue:
Product announcements
Computer drafting
Tim McCarver on Mike Bordick
Product announcements
All of the major features of version 8 are complete, and we're now focusing
on testing, bug-fixing, polishing the user interface, and finishing the
help system. We expect to begin shipping sometime in November, and we're
taking advance orders now.
We're also getting started on the 2000 Season Disk, which we expect
to begin shipping on or before December 14th. We'll be shipping that season
disk in both version 7 and version 8 format, and we're taking advance
orders for this product now as well.
Finally, we are planning to release the 2001 Projection Disk in March,
but we won't be putting that item on our order forms or on our web store
until January.
Starting today, and for the next two weeks, we are sending a mailing
to all owners of prior version of the game and to anyone else who has
contacted us in the past three years. That mailing includes a version
8 brochure, an order form, and a postage-paid reply envelope for anyone
who wishes to order by mail.
If you wish to order with a credit card, you can do so in three ways.
The most convenient method is to use our online store (follow the link
from www.diamond-mind.com),
but you can also send us a credit card order by mail or call us at 800-400-4803
during our normal business hours (9-5 eastern, Mon-Fri).
Our web site now contains a lot of information about version 8, including
a list of enhancements, a guided tour with screen shots, and answers to
the most common questions we've received or expect to receive about product
features, ship dates, and company policies.
PLEASE take advantage of those resources. We've tried very hard to anticipate
as many of your questions as possible, and we pledge to keep the site
updated with new questions and answers as they arise. We think you'll
find this to be the best way to get answers quickly and precisely. Of
course, if that doesn't help you, contact us by phone or email and we'll
be happy to help.
For a first-time buyer, version 8 will be available at the same price
as version 7 -- $49.95 for the game with 12 great teams, or $69.95 for
the game with 12 great teams plus the current season disk (2000).
Anyone who is a registered owner of any prior version of Diamond Mind
Baseball can upgrade to version 8 for $29.95. This upgrade includes 12
great teams of the past, but does not include any other season disks.
The price of the 2000 Season Disk is $29.95 in either version 7 or version
8 format. If you order an upgrade to version 8, we'll send the 2000 Season
Disk in that format. Otherwise, we'll send it in version 7 format.
We introduced our free/discounted upgrade policy on July 30, 1999. If
you bought the version 7 game since that date, we will automatically send
you a free upgrade to version 8 as soon as it is ready. If you upgraded
to version 7 since that date, you will receive a $15 discount on your
version 8 upgrade order. The mailing includes coupons that notify you
of your eligibility for these free or discounted upgrades.
Any season disks in version 5, 6 or 7 format will automatically be converted
to the format used for version 8, so your investment in season disks purchased
from Diamond Mind will retain its value.
If you have season disks in older formats, one option is to convert
them to version 5-7 first and then to version 8. If you don't have version
5, 6 or 7, we'll be offering inexpensive upgrades to those seasons when
version 8 is released. We haven't yet set the price or worked out the
procedure for these season disk upgrades yet.
If you are playing in a league that transfers statistics and manager
profiles, you can convert mid-season if every league member converts to
version 8 at the same time. It will not be possible to transfer stats,
rosters and manager profiles between different versions of the game.
Version 8 will be shipping on CD-ROM only. We can continue to send season
disks via e-mail and on diskette, but all new games and upgrades will
be shipped on CD-ROM.
We will continue to support and sell version 7 until demand for it dwindles
to a low level. If you choose not to upgrade to version 8 right away,
you can continue to purchase season disks (including the 2000 Season Disk)
in version 7 format.
Computer drafting
The title reads computer drafting, and while that's clearly the most
important new capability in the version 8 drafting system, there's really
more to it than just that. We've created a drafting environment that allows
you to:
- draft new rosters from scratch or start with partially complete rosters
(such as a group of players carried over from a previous season)
- save a draft in progress and resume it later
- browse lists of free agent batters and pitchers that show many more
stats and ratings than before, sort those lists on any statistic or rating,
and filter those lists based on batting and throwing hand, position, playing
time and more
- add supplemental draft picks for one or more teams in any round, or
delete any draft pick (to punish teams for breaking league rules, for
example)
- trade draft picks
- reverse a draft pick that was made in error
- ask the computer to make one or more picks automatically
- generate a report showing all of the picks that have been made in
a draft
These might seem like the minimal set of requirements, but we found
that they're not always present in the drafting systems in other baseball
games.
Before I get into a more detailed description of the drafting process,
let me take a moment to point out that you can still sign free agents
individually without setting up a formal draft, simply by choosing the
'Sign' command from the roster window. But if you want to set up a formal
draft, here's how you go about it.
First, you create a draft, specifying the league or leagues that are
to participate, how many rounds you want it to last, and whether the teams
draft in the same order each round (1, 2, 3, ... 1, 2, 3, ...) or in reverse
order every other round (1, 2, 3, ... 3, 2, 1, 1, 2, 3, ...).
When you do this, the full list of draft picks is generated. At this
point, you can delete picks, add supplemental picks, and trade picks.
If you change your mind about the number of rounds, you can add or delete
rounds as well. You can designate which teams will use a human general
manager to make its picks and which teams will have the computer GM make
the decisions. And you can manually change the order in which the teams
will draft or ask to have a random draft order selected for you.
Once you're happy with these things, you can freeze the draft order
and get ready to start choosing players. At this point, you can still
add, delete and trade picks (though you cannot delete or trade a pick
that has already been made) to reflect events that occur during the draft.
A human general manager selects a player from the lists of free agent
batters and pitchers. These lists use our new reporting system to provide
a powerful sorting and filtering capability that makes it easy to focus
your search on the types of players you're most interested in. And there's
a ton of data at your fingertips to help you make your decisions.
But you can also invoke the computer drafting feature to make one pick,
make all picks up until the next team that has a human GM, make all picks
through the end of a round, or make all picks through the end of the draft.
As each pick is made, the name of the player is immediately added to the
draft pick report, allowing you to follow along.
When evaluating the available free agents, the computer GM looks at
the needs of the team to make sure that it fills all of the roles (starting
pitchers, relievers, players rated at each defensive position, and so
on). And it looks at the ability of each player to help the team win in
ways that go beyond hitting and pitching to include things like defense
(range, error rates, throwing, holding runners), speed and bunting.
All hitters and pitchers are judged on a park- and league-neutral basis
so the true ability of the player comes through. This is appropriate because
Diamond Mind Baseball makes these adjustments when you play the games,
and it doesn't make sense to ignore them when choosing talent in the first
place.
If, for instance, a pitcher was effective despite facing the DH and/or
playing in a hitters park in real life, he may well be chosen higher in
the draft than someone whose stats look better but who had a more favorable
real-life environment in which to do his work.
And you might see a slugging first baseman drop lower in the draft than
you might expect. Why? Because first basemen are expected to produce more
offense than players at other positions, so it's harder to stand out.
And because some 1Bs don't supply the other things -- defense, speed,
bunting, the ability to play more than one position -- that make a player
more valuable to a team.
We believe we have come up with a pretty comprehensive way to evaluate
the overall contribution that these players make to a team, and for that
reason, we think the computer GM will provide a worthy companion in your
drafts, whether you're trying to out-think it or whether you're asking
for its help in choosing players for your team.
As each pick is made by the computer manager, it also generates a new
manager profile that includes this player. That way, when you view the
roster for this team (which you can do at any time during or after the
draft), the manager profile is already there to give you an idea of what
role the computer has assigned to each player on the roster.
As part of our work on the drafting system, we've also overhauled the
manager profile generator. Those of you who have played previous version
of Diamond Mind Baseball know that the manager profile generator was designed
to match real-life playing time as closely as possible. This doesn't make
as much sense for newly-drafted rosters, since you may draft a player
who played more than one position with the idea of using him in a different
way than he was used in real life. The new manager profile generator looks
at the talent on your roster, considers a variety of ways those players
can be combined, and chooses the arrangement that produces the best chance
to win.
As with any system that attempts to emulate the process that human GMs
go through in something as complex as building a baseball team, there's
more we can do in the future to make the computer GM even more sophisticated.
At present, it has no way to know which players are considered to be future
stars despite mediocre-to-awful stats. So it relies only on current year
performance. Some way to assess future value is one of several ideas that
are on our list of future enhancements, but even without these improvements,
we think the current version will give you a run for your money.
Tim McCarver on Mike Bordick
Yawn. Oh, sorry. It's just that I'm a little tired. You see, I stayed
up and watched the last two World Series games until they ended in the
wee hours of the morning, and I've been living in a state of sleep deprivation
for the past three days. It didn't help that I also stayed up to watch
the end of the Jets-Dolphins game Monday night, the one that broke the
record for the longest Monday Night Football game ever. (When that game
was approaching the four-hour mark, Dennis Miller joked that it was just
like a baseball game.)
But this isn't meant to be a diatribe on the length of these post-season
games, though I do think it's a serious problem when even the most avid
adult fans need to abandon a game for the sake of being able to function
the next day.
Instead, it's about Tim McCarver's game-three comment that Mike Bordick
"is not a good baserunner."
As it happens, we started working on the baserunning ratings for the
2000 Season Disk yesterday. As long-time readers know, we purchase play-by-play
and pitch-by-pitch data from STATS Inc. and analyze those data using software
that we developed for evaluating various aspects of player performance.
Our baserunning program examines how often runners took an extra base
on singles, doubles and fly balls, with adjustments for such things as
the location of the batted ball, the number of outs, whether the playing
surface was grass or turf, and whether the runner's progress was impeded
by other runners that were ahead of him.
Well, it turns out that McCarver's comment was partly right and partly
wrong. It's true that Bordick's baserunning results have been average
to below-average for the past few seasons. On the other hand, he was very
successful this year, going from first to third on singles 80% of the
time and scoring from second on singles at almost the same rate. Both
of these rates compare very favorably with the league averages. And he
did this without once getting thrown out trying to stretch a hit or fly
ball. That's a picture of a very good season on the bases.
Over the years, I've heard many comments from baseball announcers that
are totally incorrect, so it's a little unfair to pick this one out as
an example. But it does serve as a reminder that it's worth looking at
the facts -- including recent facts -- before accepting what baseball
commentators say about a player's performance. That's what we try to do
when assigning player ratings for our season disks.
|