Game 5 – Bloomquist 1 run RF
Game 4 – Aviles 2 runs @ SS (the play he threw behind Swisher @ 2B and Swisher subsequently scored w/ 2 outs, and the laser hit to him in the 1st inning that he tried to play off his hip)
Game 4 – Jacobs 1 run the laser he ole’d
Game 6 – This is debateable…I think the runs score w/ or w/o Jacob’s error in the 7th, but it is possible that he gets an out there, and Aviles plays in w/ 1 out and either cuts down (I think it was Matsui) the runner on 3rd, or holds him there alltogether…either way, I guess the play should have been made by Jacobs, whether or not it cost a run is debateable…

As for runs saved…

DeJesus’s diving catch in Left on opening day – did it save a run or not? I can’t recall…
Coco’s long run & catch in game 2 (or was that game 3) in deep right center saved I think 2 runs

i.e. I’m not counting Alex Gordon’s diving play in game 4 because it opened up the inning…Now we all know that runner on 1st and nobody out is a markedly different situation than none on & 1 out, but since it didn’t DIRECTLY take away a run…..should it count, or count in a seperate ‘baserunner saved’ category?

We just need to use our sensibilities collectively, I think, to get the most “accurate” result. I think we need to take into consideration range and arm, as well. Not necessarily just diving for plays, assists – but making reasonably difficult plays look easy. I hope that makes sense.

would be to participate in Tango’s Fan Scouting Report if/when he gets it going this season. It sounds like this gets more at what you want, and doesn’t mess with trying to convert to runs, which, for reasons I’ll allude to below, will be extremely difficult by just watching.

I’m not sure how this is going to work — it sounds like you’re grading the defense by how many runs it actually costs in the game, which makes some sense intuitively, except that doesn’t really necessarily judge how good a defender is, just how “clutch” he is. If Mike Jacobs fails to field 5 grounders that an average 1B would get to gets on first, but each time the pitcher strikes out everyone else and it costs no runs, but (for example) Billy Butler has to make plays on 5 similar grounders and makes for of them, but the one guy who gets to first gets driven on a jack, does that mean Billy Butler is one run “worse” than Mike Jacobs over those 5 grounders each had to deal with?

Maybe I’m misunderstanding what you’re doing. Defense metrics may be far from perfect (although they seem to have a better scouting eye than the geniuses who want Jacobs to play first everyday), but if this is how you are doing it… well, I’d advise reading a bit more. This is not how they work, nor should they. Doing it by how many runs (the observers think) each play actually cost on the scoreboard is to defensive metrics is like judging a players offense by his RBI. Which is about as genius as judging a player solely by what he did in 2005.

If you just judge by the plays guys “should” or “shouldn’t” be able to make, you need to have some sense of average — otherwise, errors will be total negatives, etc. Every player makes errors, fails to make plays, etc. at some point. The point of a plus/minus runs system is to compare to average.

Having said that, the idea is intriguing if you get past that. Some suggestions

1. Don’t deal with the “Run value” of events at all — just deal with the plays made/not made. I’d deal with range, mainly, as that’s the greater bulk of defense (unless you think that all the best sabermetricians are totally wrong about that, and that, e.g., the 3-1 is worth 2 runs a game or something).

2. Range is the toughest thing to judge, anyway, and the one the metrics were designed for. I’m not sure quantifying it will be that easy — one has to have a sense of what “average” is, and if you’re just watching the same guys every day (as Royals fans, that makes sense), well… that’s how people come to believe Derek Jeter, Ross Gload, etc. are good defenders. A “subjective” baseline is a tough thing to do…. Maybe instead of dealing with runs, just give each player a 1-10 or 1-100 rating or something,. but that gets more to the subjective nature of a scouting report, which I guess you don’t want to do…

3. Contrary to some, OF “arm” ratings (not as a subjective judge of strength of arm, but rather whether an OF holds/throws out baserunners) is relatively easy to calculate, but the “subjective” side of it, at least quantifying it, will be harder by just watching, since it’s hard to tell when the guy is holding the runner or not.

4. Double plays are easy enough, I guess. Again, you have to have a baseline. If you don’ t have a sense of “average” rate, almost everyone’s going to end up negative.

5. Errors are also relatively small part of defense, believe it or not, relative to range.

ONE VERY IMPORTANT POINT: No matter how you decide to work it out, the “scouts” engaged in this must not look at other defensive statistics (and who would, right? They’re all totally flawed anyway!) or read scouts opinions of these players during the season. I’m serious. You don’t want to be biased by metrics or other scouts opinions. Tango says somewhere he wouldn’t want his scouts looking at defensive numbers, since the scouts job is to see other things, and heck, one of the premises of this idea, as I understand it, is that defenisive metrics are too flawed and the fans need to correct them — so don’t be contaminated.

All kidding aside, this is a cool idea. These are just some suggestions off the top of my head. I’ll be interested to see where it leads.