Bowling stats in ODIs are so hard to make proper conclusions about. Even compared to Test stats, which are flawed at the best of times, number-cruching for the pyjama cricket leaves a lot to be desired.
The value of a wicket and a maiden changes in Test cricket depending on the match situation, but those are relatively rare circumstances, and a bowler's primary job is always to take wickets. Not so in one-dayers, where the changes happen at the same time in every match.
I wish we could see strike rates and economy rates for each period of the innings.
These bowlers will have two jobs - take wickets at the top of the innings and save runs at the death. Batsmen will be naturally more keen to play themselves in and set up a platform against the new ball, and in the last half a dozen overs wickets are bound to fall in pursuit of quick runs.
If a bowler - particularly these bowlers - takes 0/14 off his first 5 overs, we can't look at that economy of 2.8 and applaud him because the oppo were probably just seeing him off. And now the change bowlers have to try and manufacture a wicket with the field spread. If he takes 4/40 off his final 5 overs, who cares about the 4-for? He's conceded eight runs an over while a few lower-order players hole out off him.
If the spells were the other way around, it's an amazing performance. Both cases go down as 4-54, an average scorebook entry, but it could be brilliant and it could be awful.
This is all before we get into the likes of trying to work out which ODIs actually matter. In Tests you say quite confidently that almost every match had the best team the selectors thought available playing and the fans cared about it. Maybe if you're a harsher man than me you'll skim over performances against minnow teams or early tours that didn't at the time count as Tests, but in both cases you're making a minor adjustment.
Not so in one-dayers. The schedules are full of matches were first team players are rested. There are post-Test matches at the end of a long tour where the drive and contest just isn't present. There are matches where potential Test targets are picked to see how they go in internationals despite the difference in format, and worst of all there are matches that include Jade Dernbach.
You could focus on World Cup games, as the only games where everyone involved genuinely gives a ****, but then half your games are against Canada or Namibia. We're losing matches at a rate of knots here and we're only left with a fraction of the data we started with.
So it's my view that, while ODI stats are a okay guideline for identifying the top echelon of players - such as everyone mentioned in the OP plus Donald - past that the specifics just aren't there. I don't think there's any real difference in a few points of strike rate or a few tenths of a run per over, not least because those two things are often cancelled out anyway.
I guess the only way to really form the best opinion on the finest ODI players is just to watch as much ODI cricket as there is, all over the world, all the time.
Maybe we wouldn't be so quick to fill buckets with filth if we knew they had a soul. Or maybe that's what they're into. Ain't no way to get inside a bucket's mind.