How good is Sanga?

[Jassy]

It's clear that Sanga has a higher frequency of getting that big high impact score. Even considering the opposition quality - if Sanga played the same % of matches vs different teams as Tendulkar (with vs India % replacing vs SL %) and the same # of innings as Tendulkar, he would have 16 big hundreds (175+ by your definition) with 9 vs "quality" opposition and 7 vs "weaker" opposition. This compares to 13 total for Tendulkar. So even when the opposition and # of innings factors are neutralized, Sanga has a higher propensity to get a big score.



Those numbers are pretty meaningless considering how India won only 18 matches (of 73 - 25%) when Sachin played in the 90s and 54 matches (of 127 - 43%) in the 2000s . He actually had the same frequency of getting a hundred in wins in the 90s as the 2000s (22%). Both India and SL were better Test teams in the 2000s, and Sachin made 9 of his 13 175+ scores in the 2000s.

That Sanga was the better Test match-winner is easy to see:

• Sanga averaged 74 in wins compared to Tendulkar's 62
• Sanga got 18 hundreds in 50 wins (36%) compared to Tendulkars 20 hundreds in 72 wins (28%)
• Sanga's hundred average in wins is 257 compared to Tendulkar's 224 (this includes not outs)
• SL has won 40% of Sanga's matches compared to 36% Ind wins for Tendulkar (Murali's influence is neutralized by the presence of Kumble and Harbhajan, and India clearly had the better batting line-up)

Dawood's posts about strocks make a million times more sense than this.

 

[harsh.ag]

Loooooooooooooooooooooooooooooooooooooooooooooopy Looooooooooooooooooooooooooooooooooooooooooooooooooooooony ...................!

 

[Migara]

Anyway, Sanga has been pretty ordinary against better bowling units for majority of his career to really get into serious comparison with the likes of Dravid or Kallis so I won't really compare him with guys who are above.

SRT was quite weak against top class bowling attacks and that is the norm for every batsman. Kallis and Dravid struggled big time against Murali if my memory serves me right. Put it another way, when Sanga and SRT played similar attacks they averaged very close to each other. Best example is the England series in 2011 where Sanga averaged 36 and SRt averaged 39 agausnt the same set of bowlers. I can point out many occasions where they played similar attacks in nin-Sc conditons and had very similar averages.

 

[Dawood Ahmad]

Of course Richard would complain about someone having a different PPP setting.

Dawood, what are your thoughts on Nick Knight and Adam Gilchrist as ODI batsmen? Or Gavin Larsen and Brett Lee as ODI bowlers?

Adam Gilchrist: A natural match winner and a leader to victory who could easily dispatch almost all types of deliveries to boundary and crack big sixes. It didn't matter weather he scored big or not, if he scored Aussies would mostly win in an unusual way, no matter what the conditions and situations were.

Nich Knight: A Hashim Amla style opening batsman who easily keeps the scoreboard moving with high speed but has countless weaknesses in middle overs and builds no innings of high qualities. Till such batsman remains on crease, pressure remains.

Update:In there comparison, things would simply go with Gilchrist due to his better attack, defense, timing, middling, stroke, persistence, foot work, shot selection, resilience in body etc.

 

[Jassy]

Sanga averaged 30 in the 2011 series, not 36. It is all irrelevant anyway for obvious reasons. Honestly, there is an argument or two you could make for Sanga like Maximas did gracefully without feeling the need to make up stats or misuse them, but when you compare him to blokes like Ponting and Tendulkar you shoot yourself in the foot and the last argument you want to be making in favour of Sanga is his stats outside the subcontinent because last I checked Ajinkya Rahane has about as many good series as Sanga against ''comparable attacks'' outside the SC.

 

[simonlee48]

Sachin has 3 times the tons in wins in the 2000s because India won 3 times as many matches (being a clearly better Test side in the 2000s than in the 90s). To correct my previous post, he did hit tons more frequently in the 2000s (30% ton rate vs 22% in the 90s).

Obviously wins are not down to one individual, and especially Test wins are heavily influenced by having quality bowlers, but the Indian bowling unit was at least as good as SL's during Sachin's career and the batting unit was mostly better so the two are more comparable. Actually if you compare Dravid to Tendulkar you will most likely see that he was more of a Test matchwinner than Tendulkar.

Let's look at both in 2000s to comapre apple to apple and not abuse stats here.

• Sachin played in 35 wins and has 12 tons in those wins
• Dravid played in 39 wins India and has 11 those wins

You are here abusing stats to prove a point which doesn't exist. If you take the entire career of SRT and career of many players from India in 2000s then you could prove that all those Indian player were bigger match winner than SRT. Same is the case of Sanga here.

As I said, anyone who has watched cricket, will never take Sanga over the likes of SRT or Lara for winning matches. It's absurd to suggest that. How do you win matches when Sanga has been ordinary against above average bowling sides for majority of his career?

Don't mix team result with individual to make a point specially when teams don't even play all oppositions in similar proportions.

 

[simonlee48]

SRT was quite weak against top class bowling attacks and that is the norm for every batsman.

Well, we are not talking about some great bowling units here. Simply above average bowling units here.

There are only 8 non-minnows. Sanga faced 7 of them. So you have 10 year period( 5 year in 2 chunks) with 52 tests against 4 above average bowling units faced by Sanga and he has been pretty ordinary against them in that period. He did well against them in middle period where he played total 11 tests.

That's not the norm for every batsman let alone SRT. It's will be absurd to put SRT/Lara/Dravid/Kallis etc and Sanga at same level when it comes to consistently performing against better bowling units in their entire career.

Just in case you missed earlier post,

---------------------------------------------------------------------------------------------------

Sanga against SA, Aus, Eng and NZ in his first 5 years: Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Sanga against SA, Aus, Eng and NZ in his last 5 years: Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

He did well against them in middle period where he played total 11 tests.

----------------------------------------------------------------------------------------------

Not that Sanga was bad or could never play good knocks but I will never bank on Sanga to consistently score runs against above average bowling units. I will bank on the likes of SRT, Lara, Kallis, Dravid and all. That's the point I was making. Now despite being ordinary against above average bowling sides for majority of career his aggregate stats are better and that's the main reason many folks started listing him with the likes of Kallis/Dravid but he is not in the same class.

 

[viriya]

Let's look at both in 2000s to comapre apple to apple and not abuse stats here.

• Sachin played in 35 wins and has 12 tons in those wins
• Dravid played in 39 wins India and has 11 those wins

Why just 2000s? Dravid averages 66 to Sachin's 62 in wins, and has 15 tons in 56 wins to 20 tons in 72 wins. It's not as if I checked the data when I made that comment. If anything, this shows that Sachin was not clearly a better matchwinner than Dravid in Tests. Either way, Sachin vs Dravid isn't the discussion here.

Sanga has not been ordinary against better bowling atttacks, your claim that Pak bowling in the 2000s was somehow comparable to Ban or Zim also doesn't make sense. Only the Aus, SA, Eng and SL attacks have had a lower overall bowling average than Pak since Sanga debuted in Tests:
Team records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Sanga averages 53 vs non Ban/Zim opposition to Tendulkar's 51.
Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo
Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Lara averaged 53, Dravid averaged 50 and Kallis averaged 53:
http://stats.espncricinfo.com/ci/en...orderby=default;template=results;type=batting
http://stats.espncricinfo.com/ci/en...orderby=default;template=results;type=batting
http://stats.espncricinfo.com/ci/en...orderby=default;template=results;type=batting

 

[simonlee48]

Why just 2000s?

This match winner logic can make some sense only if other factors are constant otherwise we are abusing stats here. 2000s was the period when Indians were winning. Dravid and SRT happen to play the full decade for the exact same team against same oppositions. There were still few instances when one played and other didn't but this is as close as you can get for doing meaningful comparison.

Talking about match winning tons for players, who played for different teams or against vastly different oppositions in their entire career, is simply abusing stats here. Team performance and individual performance are not the same thing. Team results are decided based on how 22 players do in any given match.

 

[viriya]

How is comparing averages in wins abusing stats?

 

[simonlee48]

How is comparing averages in wins abusing stats?

You were trying to compare Sachin's entire career's tons in wins with Dravid's or Sanga's to make a misleading point. Winning game is team result which is based on how 22 players do.. They never played for the same team and against same oppositions. To still entertain you, I even compared 2000s for SRT and Dravid because they payed the entire decade for the same team and had almost similar oppositions. They pretty much participated in games where same 22 players were playing. What did it show? Did it show that David was making more tons in wins? Team achievements and individual achievement shouldn't be mixed when comparing players. We are not comparing two teams here.

If you still didn't get it, I can't really explain to you.

 

[viriya]

My question was:

How is comparing averages in wins abusing stats?

not tons in wins in 2000s vs 90s.

 

[simonlee48]

My question was:

How is comparing averages in wins abusing stats?
.

Win is a team performance which depends on how 22 players do in any given match. Avg/runs/tons in wins ( a subset of players entire career which depends on team's performance) are not comparable unless you are you are playing for the same team and against the same oppositions.

You can compare team performances or you can compare individual performances but mixing them to make a point is called abusing the stats. I hope it's clear enough now.

 

[viriya]

So the simplest stat like average is "abusing stats". But picking and choosing what opposition is worthy (a la Pak bowling attack) is accurate statistical analysis? Got it.

It's as if a batsman has no impact on a team's wins. If two sides have similar strength bowling attacks then it's not unfair imo. Either way, your average in wins is independent of how many wins you had.

 

[simonlee48]

So the simplest stat like average is "abusing stats".

Abuse of stats starts when you take a subset of individual stats based on team's stats. Neither Lara was less capable of winning matches nor Sanga is more capable of winning matches because runs/tons/avg in their team wins says so. Assume player A can play consistently well against above bowling sides. Another player B has trouble playing well against above average bowling sides. You can have 10 fixed players and form a team of 11 players by picking either player A or player B.

Do you seriously think that player B is going to help his team win more matches? That will be devoid of any logic. Player B here is Sanga and Player A here is SRT, Lara, Dravid, Kallis, Ponting etc.

But picking and choosing what opposition is worthy (a la Pak bowling attack) is accurate statistical analysis?

You can simply see how each team's bowling unit has done. You are not mixing individual stats with team stats. You are only comparing Pakistani team with others. No issue in doing that. As long as you stick to comparing teams performance with each other or individual performance with each other, it make sense.

It's as if a batsman has no impact on a team's wins. .

A batsman has a chance to make an impact but not different than other 21 players. 22 players collectively decide who will win the game. Individual performance of a batsman in won games hardly tells you anything about the quality of any batsman. Remember he has less than 5% influence in any game so no point in taking some random matches where his team won and then drawing any conclusion.

On other hands, individual performance of a batsman in all games tells you a lot about that batsman.

 

[viriya]

Except that Sanga doesn't have a problem vs good bowling sides (your definition that excludes Pak hardly makes sense as I pointed out earlier, convenient since Sanga has dominated them unlike anyone else - Ban/Zim I can agree with).

Pak has a better overall average than NZ, Ind, WI, Ban and Zim, and it's hardly any different from how SA, Eng and SL bowling attacks have done since Sanga's debut. On the basis of this you can probably argue for excluding WI, but not Pak:
http://stats.espncricinfo.com/ci/en...pan;team_view=bowl;template=results;type=team

I was comparing Sanga and Tendulkar - not Lara/Kallis etc. The reason they were more comparable when it came to win records was because the bowling units of Ind and SL were similar in strength and Ind batting had an edge. Tendulkar played most of his matches during the same time as Sanga and their batting average difference in wins is more than 10 runs (Sanga 74 vs Tendulkar 62). You would need to "abuse stats" to claim that Sanga was not a better Test matchwinner than Tendulkar.

Just because a player is just 1 of 22 players in the game does not make his influence 5%.. Did Murali have a 5% influence in the games he played? Of course not. Some players influence games more than others. Consider Inzamam. He averages 50 overall, but 78 in Pak wins - a clear example of a matchwinner.
http://stats.espncricinfo.com/ci/en...2=runs;result=1;template=results;type=batting

 

[simonlee48]

Except that Sanga doesn't have a problem vs good bowling sides (your definition that excludes Pak hardly makes sense as I pointed out earlier, convenient since Sanga has dominated them unlike anyone else - Ban/Zim I can agree with).

Pak has a better overall average than NZ, Ind, WI, Ban and Zim, and it's hardly any different from how SA, Eng and SL bowling attacks have done since Sanga's debut. On the basis of this you can probably argue for excluding WI, but not Pak:
Team records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

I was comparing Sanga and Tendulkar - not Lara/Kallis etc. The reason they were more comparable when it came to win records was because the bowling units of Ind and SL were similar in strength and Ind batting had an edge. Tendulkar played most of his matches during the same time as Sanga and their batting average difference in wins is more than 10 runs (Sanga 74 vs Tendulkar 62). You would need to "abuse stats" to claim that Sanga was not a better Test matchwinner than Tendulkar.

Just because a player is just 1 of 22 players in the game does not make his influence 5%.. Did Murali have a 5% influence in the games he played? Of course not. Some players influence games more than others. Consider Inzamam. He averages 50 overall, but 78 in Pak wins - a clear example of a matchwinner.
Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Arr, SRT started in 89 and Saga in 2000. SRT's peak was past it before Sanga started. When did you start watching cricket? Genuine question because match winner stats may show you that SRT was crap in 90s and great in 2000s but trust me it's not true. I saw pretty much all games, where he played, after his first knock at Perth.

We already saw that Pakistan has been the easiest opposition to score against in their home in the last 10 years. Their flat pitches contributed a lot to that in my opinion. Credit to him for scoring runs against Pakistan because Pakistan is not really Zim or BD. But he has been pretty ordinary against SA. Aus, Eng and NZ for majority of his career. I do think that all 4 have better bowing units than Pakistan and that's why I said his performance against above average bowling units has not been good for majority of his career. He played only 7 non-minnows and here you have 4 of them.

I have followed Sanga's career and I was using stats only to show what I always felt. No point going in circles. Let's drop it and agree to disagree.

 

[viriya]

His peak was late 90s.. his best Test years were late 90s-2000s.. I started watching mid-90s.

 

[simonlee48]

His peak was late 90s.. his best Test years were late 90s-2000s.. I started watching mid-90s.

Nope his best decade was entire 90s by some margin when you are talking about his test career. I forgot exactly when he had some injury but he changed his batting style and started playing less strokes in 2000s. Not the same batsman after that.

See how many batsmen actually had good time in 90s and where he stood among them in 90s, Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

See how may bastmen had good time in 2000s and where he stood among them in 2000s : Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Don't go by number of matches. He was a different payer in 90s from early days in the test format.

 

[viriya]

Like I said his first peak was late 90s.. if you do 1990-95 it's not as good. Lara does better:
Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

He was better to watch in the late 90s, but was as effective if not more in the 2000s (barring the lean patches). I value results over how it looks myself.