Averages just tell you how many runs someone scored for every time they got out. They can't be misleading. Unless they're wrong.
The average is simply a mathematical number. It is not misleading by itself. The interpretation is what is at times misleading.
Take Mohammad Yousuf.
His average is 53.
During the recent Pak-Aus series, the Aussie commentators kept saying how he is one of the best batsmen in the world, and he is so good, and they kept mentioning that he averages 54 (he did at the start of the series). Shane Warne kept saying "anyone who averages 54 is a dam good player" On the other hand, a Chris Gayle, who does not average 54, and is not considered one of the best batsmen in the world blasted two centuries against the same attack against whom Yousuf failed so miserably.
Now I am not trying to say they are talking crap. What I am saying is, they are obviously being diplomatic and nice to talk up the competition and using his average of 54 to back themselves up. That is the misleading bit. It is misleading because it is incomplete.
Mohammad Yousuf started playing in 98, and I was already an avid follower of cricket by then. So I have followed his career all through. For most part, until 2006, he was never considered in the big league of the best batsmen of his time. He was seen as someone who times the ball well, is stylish, looks very good at times, will score a brisk 40-50 and get out. That was used as a criticism and a reason why he is not considered in the big league.
Then came 2005, when Pakistan started playing a lot of cricket at home. They played England, India, West Indies. Yousuf scored heavily against these teams, and on pitches that were flat as anything. Some of the totals notched up in these matches will tell you how flat the pitches were. Needless to say, neither of these 3 teams had a great bowler at that time. Harmisson was useless on dead pitches. Flintoff was decent at best.
It was during this time that he broke Vivian Richards' record in 2006, and all of a sudden people started taking notice. I think he scored 9 centuries and his heavy scored boosted his average. At the end of the year, Pakistan toured South Africa for 3 test matches, and Yousuf was in the form of his life..but considering how South Africa had a slightly better bowling attack than what he had faced all year, he was back to being the old Yousuf. He scored two 80s in the 3 match series that Pakistan lost 1-2.
The point is, his average of 53 places him as an equal of Lara, Tendulkar, Kallis, Ponting when it is clear he is quite below them. I am sure with more test matches against quality opponents, his average will fall more and maybe at the end of his career, his average will be 49-50.
So once again, it is not the number, it is the interpretation of that number which is misleading. I am not saying stats are crap. But to say Mohammad Yousuf average 53 is providing incomplete information. It should also add that he averages 30 and 29 against two of the best teams in the world ( Australia and South Africa). I am using stats to make my point, just adding a factor to put things in perspective.
I still remember a test match against India in 2004, where Sehwag got his 300. Pakistan were following on and the match had already gone out of their hands. Yousuf had failed in the first innings, and in the second innings, score a brisk but completely inconsequential century, and unlike Sachin's 100 in the first test against SA, his 100 came when it was clear Pakistan were going to suffer an innings defeat.
The commentators Sanjay Manjrekar and Imran Khan remarked how he could not play the same innings in the first innings against the same attack when there was a bit more pressure on him, and how Yousuf had so far (2004) managed to fail when put under pressure, but bat freely when there is nothing to lose. So that 100 will obviously boost his average and runs, and rightly so, because he did get that a 100, but we would be misleading ourselves if we fail to acknowledge the context.
Another prime example is Thilan Samaraweera. Cricinfo writer Sambit Bal had an excellent article titled "how good is Samaraweera" where he touches on these issues of "misleading statistics" and argues well to put forward his point, that Samaraweera, despite an average of 50 plus is not all that. He uses statistics to make his point too, because statistics by itself is not misleading, but incomplete statistics is.
Say a bowler playing 50 Tests had 1 dropped catch in every match those possible 50 wickets would alter his averages.
