Some quick tests with Hugs +s on a example list that I constructed with 576 elements give food for thought:
reductions cells my one liner 4035 11483 tournament 7053 12288 your penultimax 16715 20180 your penultimax2 7466 10344 your penultimax3 8605 13782
And with GHC -O, the results are quite different: Mark's one liner 0.60s tournament 0.25s your penultimax 0.26s your penultimax2 0.25s your penultimax3 0.25s the tests were with random numbers (different seed each time), but there was only a little little fluctuation from run to run. I specialised all the algorithms to Int. Hugs doesn't do much optimisation, so the reductions/cells counts tend to vary quite a bit between programs which GHC will optimise to the same or similar code. The moral, as usual, is YMMV... Cheers, Simon