But that's my point. It's possible to quantify how much a dropped catch costs - it's around the average of the batsman dropped. Let's say that against an ATG team a dropped catch is worth 45 runs. This is pretty generous - 44% of batsmen who make a hundred will have a chance spilled by then (see article linked below).
Now Gilchrist averages 40.6 runs per innings or 58 runs per test with the bat. Knott averages 29.5 runs per innings or 46.2 runs per test. That's a gap of 11 runs per innings or 12 runs per test. Note that the second comparison is extremely unfair on Gilchrist because he batting in less innings than Knott. So let's go by the difference per innings since in a theoretical all time side Gilly and Knott are going to bat in roughly the same number of innings. So in 4 innings Gilly scores 48 more runs than Knott. If Knott takes one catch that Gilly would have dropped every two tests then they come out even. But the difference between the keepers probably isn't that great. How many catches does a keeper drop in a test on average?
According to
Tracking the misses | The Cricket Monthly | ESPN Cricinfo Gilchrist missed 12% of chances during the 2003-2009 period. Let's pretend that's the entirety of his career. Now let's also pretend that Knott was so good as to miss a mere 5% of chances (half that of Boucher, the best in the period at 10%).
Gilchrist in his career took 4.3 dismissals per match. With a 12% drop rate he dropped one batsman roughly every second match or 50 chances.
Knott took 3.0 dismissals per match over virtually the same number of tests. With a 5% drop rate, Knott dropped 13 chances in his career.
So Gilchrist dropped one chance every other match while Knott dropped one chance every six or seven matches (14% of the time). Now that's fantastic. Surely that proves that Gilchrist is worse for the team? Not necessarily. Gilchrist took nearly 50% more dismissals than Knott. If we extrapolate and suggest that in an all time side, Knott gets the same number of opportunities as Gilchrist (466 chances total if Gilchrist spilled 12% of chances in 96 tests), Knott then missed 23 chances in his career. So Knott take 27 more chances over the course of his career. 27 chances multiplied by 45 runs tells us that Gilchrist cost his side 1215 more runs over the course of his career than Knott would have.
As we established earlier, Gilchrist was worth 11 more runs per innings than Knott. If they batted the same number of innings (let's take Gilchrist's figure of 137 as the example), that means that Gilchrist would have scored 1507 runs more than Knott over the course of his career. That means that by playing Gilchrist over Knott you get a bonus of 1507-1215=292 runs during the course of his career.
Now I've been quite generous to Knott here on a number of counts. I'm assuming that Knott took nearly 60% of the chances that Gilly spilled. He would have had to be extraordinarily better than Gilchrist for that to be the case and the stats suggest he wasn't. The gap between their WPM figures suggests that Knott wasn't substantially better than Gilchrist as a keeper. Secondly, I've used the runs per innings rather than the gap between averages to measure the difference between the two with the bat. Thirdly I've assumed a ridiculously high figure of 45 runs per innings scored on average after a batsman has been dropped. The lower this number actually is, the less wicket keeping perfection matters. Finally, the linked article actually mentions that Gilchrist was below 10% for a substantial portion of his career. Keepers get worse towards the end of their careers. It's likely that GIlchrist spilled substantially less than 50 chances over the course of his career.
The point is that the better a keeper gets, the less their keeping ability matters compared to their ability to score runs. If Gilchrist spilled 12% of chances (including difficult ones), a theoretically perfect keeper would still need to average 25 runs per innings to be worth more to the team than him over the course of his career (assuming the average batsman makes 45 runs per innings after being dropped).
The ability for bowlers to create chances is far more important than the keeper's ability to take chances once the keeper is good enough.
Phew, this has been really stats heavy, but I was interested in discovering just how important keeping ability is. As I suspected, it's not quite as important as it seems for the top glovesmen.
So just remember kids, a wicket keeper who never drops a single catch in their career still has to score 24 RPI (average at least around 30) to be as valuable to the team as Gilchrist.