One thing that people don't understand about computer chess is that computers get weaker and weaker as the time controls increase. You might think that having a day to pick your move would benifit the computer which can calculate millions of positions per second. But as a matter of fact it does'nt.
On fritz the branching factor is roughly 3 to 4 branches per ply. That means that to get one ply further in the game tree, fritz needs to calculate for about 3.8 times as long. So.. take the typical search.. In about 10 seconds or so fritz will have searched about 12-ply. This is a hell of a lot more than a human will have calculated in 10 seconds.
by 40 seconds fritz will be at 13-ply.. probably still more than a human.. but not more than a strong human..
by 2 minutes it will be at 14-ply.. by no means further than your average master..
after a day.. it will only be at abuot.. 20 ply.. and by then a human will have gone much much further than 10 moves ahead (which is 20 ply). not considering every possibility like fritz will have.. (save for those lines pruned away by alpha-beta and mtd(f) and whatever algorithms are in play), but the important ones to be sure. And the human has a much higher quality evaluation function.
20 ( +1 | -1 ) Wellconsidering computers are allowed in most correspondence chess organizations, I would imagine he would have beat Fritz 8 easily enough, since most of his opponents probably used it.