since the 2007 series against west indies the only teams they've beaten at home are bangladesh and sri lanka recently, and even the sri lanka series contained a hiccup. my feeling is that south africa look the best team on paper, and have for a while, but they've still to show they're more deserving no.1s than india or england.
They drew at home to England, Australia and India, their only home loss in said period to Australia 1-2.
England got "murdered" 1-1 in South Africa, holding on desperately in both the 1st and 3rd Test. You can praise England's rearguard as much as you like, won't change the fact the South Africans were better in 3 out of 4 Tests.
Australia won 2-1, all three tosses won by Australia and their two wins both batting first. South Africa managed to bowl them out cheaply in the 1st innings once the series was already lost and won by an innings.
India got a 1-1 draw thanks to two batting failures by South Africa, how often would you expect them to be bowled out for 131 and 215 in the same Test? Zahir and Harbhajan taking six wickets each
The more recent 1-1 draw against Australia was just two Tests, the 1st Test was just silly with Australia skittled for 47 and South Africa 96 before South Africa took control in their 2nd innings to score 236/2 to win. Australia won the 2nd Test, a tight encounter not entirely dissimilar to this Test just finished.
You want to look at England's home record, we've lost to South Africa at home twice and India once.
Home (2007-present)
South Africa P8 W4 D3 L1
England P12 W9 D0 L3
England get through about two series a summer, won 75% but three losses to 'rivals for the top spot'. Six of those nine wins were against West Indies, Bangladesh, New Zealand or Sri Lanka, three against West Indies - South Africa have won all four of their series played against the same group of sides. Against the rest England's record is won three, lost three
I'd say the key difference is length of series. England play the top sides in 3+ match series, South Africa have played just two four match series since 04/05, BOTH against England (won away, drew at home) England's three wins over better sides have come in 4-5 match series, as have their defeats. Conclusion? More wins come when playing 4+ match series, 2-3 match series are more vulnerable to be drawn.
And maybe the main conclusion is if South Africa stopped poncing about with short Test series they'd have been top of the pile a while back.
----------
I'd suggest we've actually been just as successful seeing as that Flintoff side also enjoyed a 2 year winning streak that culminated in an Ashes win. Much like the current team did before going to Asia and getting spanked. The similarities are actually incredibly similar, with the only difference being Australia are now an easier team to beat and had given up the massive lead in ranking points they had back then.
We had a purple patch with five bowlers against South Africa and Australia in 04-05 to 2005, mainly because Flintoff had two big series with bat and ball, and Jones did well with the ball, as did other quicks - Hoggard 26 wickets in South Africa. Those series aside we picked off weak sides and had an indifferent record. Even against the aussies we were far from convincing, wins by two runs and three wickets in a 2-1 series win. Without Giles we probably would still have won both series, maybe more convincingly.
Flintoff was very hit and miss, never enjoyed the same success as those two series ever again. Flintoff only ever took more than 14 wickets in a series in those two series, he enjoyed some good series with the bat
04/05 vs South Africa
Flintoff - 227 runs @ 28.38 & 23 wkts @ 24.96
Jones - 15 wkts @ 26.67
2005 vs Australia
Flintoff - 402 runs @ 40.20 & 24 wkts @ 27.29
Jones - 18 wkts @ 21.00
We also got THRASHED 0-5 down under in 06/07, you can cite all the rare examples in the world but for every example of a five bowler series win there's a loss. I've done the analysis, while we might have gotten away with it at home we've done far better overall and away with four bowlers.
Since 2000
Five bowlers : P82 W34 D22 L26 (Won 41.46%, Lost 31.71%) 20W = 47.56%
Four bowlers : P79 W39 D20 L20 (Won 49.37%, Lost 25.32%) 20W = 53.16%
Won more Tests in less played with one fewer bowler
Home
Five bowlers : P41 W24 D9 L8 (Won 58.54%, Lost 19.51%) 20W = 65.85%
Four bowlers : P48 W26 D12 L10 (Won 54.17%, Lost 20.83%) 20W = 54.17%
Won a slightly better percentage of games with five bowlers, but barely significantly.
Away
Five bowlers : P41 W10 D13 L18 (Won 24.39%, Lost 43.90%) 20W = 29.27%
Four bowlers : P31 W13 D8 L10 (Won 41.94%, Lost 32.26%) 20W = 51.61%
20W is how often we take 20 wickets, an obvious thing to analyse as some think playing five bowlers somehow increases your chances of winning and to win you usually need to take 20 wickets. If you get bowled out twice for lack of batting then your chances of bowling twice are reduced.
There is a clear case, if one were needed beyond theory, to play four bowlers. We don't have an all-rounder capable of batting 6/7 so the argument is rather lame for, I would despair if people seriously 'think' Bresnan, Broad and Swann could provide adequate compensation for a proper batsman. If we'd played five bowlers in the final Test we'd have probably lost much more quickly, for starters you'd have to find the runs Bairstow scored as he'd be the one left out - around 150.
EDIT : The major flaw with arguing five bowlers based on past success is of course we don't have Flintoff, part of whatever success we have had with five bowlers.