I went to an afternoon baseball game Thursday. The temperature was 72 degrees and the sky was cloudless. But despite looking like a giant mall from the outside, the ballpark has surprisingly few entrances so there's a lot of walking in crowds to get in. I suppose it's not much worse than Dodger Stadium, but since that's built into a hillside it's less obvious there. Once I got inside, the sightlines were lousy because the seat angle is too shallow. You have to crane to see the plate, and the staircase railings are intrusive (although they do use glass to minimize it, that's a bandaid after the fact).

So I was grumpy, despite all the suntan lotion seeping into my pores, and the only Dodger paraphanalia I had was my pencil, which needed sharpening. I was pretty much in a funk from the third inning to about the eighth inning, when my mood, very surprisingly, lifted considerably, and from then on I had a lovely time. (Correlating this information with the box score is left as an exercise to the reader.)

How did the new scorecard do? See for yourself: Front side, back.

Success

• The light background lines work really well. They are visible as guides but still keep the pencil marks most prominent.
• The pitch count area is a good graphical representation; the difference between pitchers is very obvious. Lawrence was on cruise control into the 8th inning, and was pulled at 83 pitches and 7+ IP only because the manager got nervous. (By rule of thumb, 100 pitches is when you start to think about pulling a pitcher, and 120+ is a danger zone for injury. For fragile pitchers, such as Pedro Martinez, the numbers are 80 and 100, but fragile, excellent pitchers can throw 9 innings in under a hundred pitches, so that doesn't reduce Pedro's value much.) Dessens was the opposite, throwing 93 pitches! in four innings. Dessens is a righty, the Padres lineup was heavy with lefties, and he appears afraid to throw inside to them. So he kept aiming for a magical spot six inches off the plate where nobody could possibly hit the ball, but only Tom Glavine gets strikes called there, so it was pretty brutal to watch. I'm stunned that Dessens only gave out four walks, and only had a single full count in twenty batters.
• The out counts in shaded, unbordered discs are nicer than the previous little circles with black borders.

Failure

• My previous scorecards had player # after player name, and position in a little box. For some reason I reversed it, and it screwed me up. I will change it back.
• Since I've stopped trying to mimic newspaper box scores, the only purpose of the player stats columns on the right is to show, at a glance, who had a good night. Having five columns, two of which are derivative, isn't helping that. I could replace OBP and SLG with OPS, but then I have to figure out an easy way to calculate that.
• I started by keeping the pitch counts on the same side as the pitchers' teams, but that's the opposite side as the batters, so after two innings I switched to the other way, which is much easier since you can just count on top and draw lines on the bottom. I think I'm going to put the pitcher stats on the same page, which means that each page will show one team's batting and the other team's pitching. But I guess that makes sense.
• The pitch count boxes can be much smaller. I went with four boxes across because I wanted to use shading to show at a glance whether or not the pitcher is staying at the magic ~12 pitches an inning for a complete game. You can see that pretty well in the first seven innings of the front side. But I ended up with a checkerboard pattern and a lot of base-twelve math. It should be easy just by looking at the lines to see the pitch count. I think I'll make it five or maybe 10 pitches across, add counters, and shade in a stairstep pattern.

I also found this site with lots of baseball scorecards. Some are intended for youth games or other participatory stuff like coaching, and I did pick up the cool idea of putting everything on one side of one (extra-large) sheet of paper. But the real problem with all of them is that they are ugly. They would all benefit from a hefty sprinkling of Tufte; specifically, they have terrible data-ink ratios. That is, they have a lot of heavy black lines and grids for a small amount of writing, especially given that the writing may be in pencil. The Mk 2 Scorecard was also guilty of this, but I believe I have corrected the problem with the Mk 3 and, after I add the refinements outlined above, will try submitting it to the aforementioned site.