Game diasection taken to a whole, nother, level.

Originally posted by Brent

   Prepare to be shocked that it is Bill Abner

Most team sports, strategically, boil down to this: creating and controlling space.

Football.
Basketball. Hockey. Soccer--creating and controlling space. So as I
discuss the train wreck, the utter ineptness, of NCAA 09, keep the idea
of space foremost, because it's critical to understanding why this game
is so embarrassingly poor.

Also, as I describe how this game is
so totally broken, I'm going to discuss data gathered by watching CPU
vs. CPU games. Once I see that something is wrong as I'm playing the
game, I'll usually confirm my observation by watching the CPU play
itself. Even if I'm playing , over 95% of the players are being
controlled by the CPU at any one time.

Here's a core requirement
for a sports game: have players run at the correct speed. Not just the
correct absolute speed, mind you, but the correct speed in relation to
each other. One of the first things I noticed as I started playing NCAA
this year was that some players seemed to be running in quicksand--they
were so disproportionately slow that it looked ridiculous.

I
decided to test what I was seeing, so I went into practice mode and
tested how fast various players could run the 40-yard dash, using the
streak route (with no defense). In all cases, the players appeared to
be sprinting with maximum effort.

The fastest player in the
game? A receiver from Florida (Percy Harvin). 98 speed, 99
acceleration. His 40 times? Between 4.6 and 4.7.

That seems a bit, um, slow.

Michael Crabtree of Texas Tech? 94 speed. Five timings. Between 4.7 and 4.8 every time.

Here are more 40 times for different speed ratings.
90 speed: 5.2 to 5.4
85 speed: 5.5 to 5.7
80 speed : 5.9 to 6.2
75 speed: 6.1 to 6.3
70 speed: 6.5 to 6.7

Acceleration
will affect those times as well as the speed rating, so I tried to find
players with proportional acceleration ratings (for instance, a 90+
speed player with acceleration of 90+ as well).

Here's some context for those numbers:
--roughly 10% of the linebackers have speed above 85
--almost no offensive lineman have speed above 70.
--nine defensive ends have speed above 85
--one defensive tackle has speed above 80

In
other words, a bunch of guys look like they're running in quicksand
because they ARE running in quicksand. Take a look at real NFL combine
40-yard times times from this year:
QB: 4.55 to 5.03
RB: 4.24 to 4.83
TE:4.53 to 5.04
WR: 4.31 to 4.76
OL: 4.98 to 5.56
DT: 4.88 to 5.49
DE: 4.57 to 5.36
LB: 4.47 to 5.12
DB: 4.31 to 4.80

It's
absolutely fair to say that combine times aren't the same as in-game
times, because guys run at the combine with track shoes on an
artificial surface. The difference shouldn't be nearly half a second,
but they're not going to be as fast.

The problem, though, is the
gap between fast and slow. The fastest offensive lineman in the combine
was .67 slower than the fastest wide receiver. In NCAA 09, that gap is
over 2 seconds for over 99% of the lineman. With defensive tackles,
it's .57 in the real world, and over 1.2 seconds for almost every
player in the game.

Think this doesn't matter? Let's look at how
these kinds of sizable inaccuracies affect the game, notably in the
area of pursuit.

It's been noted (and I've observed) that
playing man-to-man defense in this game is usually a deathwish. It's
also been noted (and I've observed this as well, repeatedly) that
quarterbacks have insanely high completion percentages (70%+ is common,
and I rarely see anyone below 60%, even with the Accuracy slider
reduced below the default setting).

Well, speed disparities are
why that's happening. A linebacker covering almost anyone is going to
be left in the dust (with rare exceptions). The speed gap is so
exaggerated that it essentially obsoletes half the defenses in the game.

The
CPU will keep calling those defenses, though, and quarterbacks will
roast them, which partially accounts for the high completion
percentages (the quarterbacks are also psychic this year, but I'll talk
about that tomorrow).

Kickoff and punt coverage is hideous this
year. Why? Speed disparities. Lineman are slogging downfield so slowly
that huge running lanes are opened. There's also a blocking issue,
which I'll discuss tomorrow, but speed is a big part of the problem.

Many
of the specific bugs I'll discuss tomorrow could, theoretically, be
fixed. The problem with speed, though, is a design failure. It's not
going to get fixed.

How often will you notice this speed issue?
All the time, and particularly on plays where defenders should have an
angle on an offensive player, but don't because they're so slow. In the
open field, most defenders have absolutely zero chance of ever making a
tackle because of the difference in speed. It's done to create more big
plays and amp up the offense.

In short: design cheese.

If
this was the first year of using a new engine, I could see having some
speed issues. Tiburon, though, will have something right, then break it
the following year. They seem to be unable to determine what works and
what is good versus what is broken.

Tomorrow I'll look at the
stunning number of gameplay issues not related to speed--basically,
every single aspect of gameplay has major issues, and we'll discuss
them using a game between two premiere teams: Florida and LSU.