Test careers that started and finished strong

How they started and how they finished: the batsmen

The idea for this article came to me when Hussey announced his retirement from International cricket towards the end of 2012. I had an instinctive feeling that Michael Hussey had a great start to his career as well as a great end. I thought it was worthwhile looking at the start and end of the careers of all players. And I was almost certain that another player, right at the top of many batting lists, would announce his retirement from Test cricket if he had an ordinary home series against Australia. Well, Sachin Tendulkar had an ordinary series but he did not announce his retirement.

However, I did not want one player's inability to take a call on his fabulous but fading career have an influence on the timing of an important article. Hence, I have done this article knowing fully well that Tendulkar is still on for at least the next series. Let me also say this. Tendulkar's last ten Tests have been played at home and he has scored 367 runs at an average of 24.46. In three away-Tests in the forthcoming Test series against South Africa, facing Dale Steyn, Morne Morkel and Vernon Philander, would he suddenly score 400-500 runs? Even the staunchest of Tendulkar's supporters would realise the futility of such expectations. It is likely that he scores only 200 to 250 runs in these three Tests. So this article might only undergo minor changes.

I am going to standardise the criteria. First, the start of a player's career is defined as the first ten Tests and the finish of his career as the last ten Tests. The cut-off for this analysis is batsmen with 4,000-or-more Test runs and thus, 115 batsmen qualify. Everton Weekes crossed 4,000 comfortably and played 48 Tests, which is the lowest any player has played in this list. So, even he has a middle period of 28 Tests. Because of the rather high cut-off, batsmen like Hanif Mohammad, Dean Jones, Arthur Morris, Tony Greig, Clem Hill et al. do not qualify. But this also means that non-batsmen such as Shane Warne, Chaminda Vaas, Wasim Akram and a host of wicket-keeper-batsmen are also excluded. And, I am very close to the 50-Test cut-off which I was aiming at.

I did not want to filter out players who are still playing Test cricket. That would exclude many players and is an artificial restriction, especially since some of them may play very few Tests in future. I am also going to use the Batting average as the measure of comparison. It is the most used and acceptable of performance measures. Runs-per-Test value implies quantity rather than performance and Runs-per-Innings metric has its own shortcomings especially as we have batsmen who have batted from no.1 to no.8.

I had initially planned to complete both batsmen and bowlers in the same article. Then, as normally happens, I ended with 8 tables and one graph for the batsmen. I did not want to test the ability of cricinfo's production team to handle 16 tables and two graphs in one article. So the article on the bowlers is the next one.

I have incorporated different types of analysis in this article. I have tables which list the best and worst starts to the player's career. I have the best and worst finishes to a player's career. Then I have combined the start and finish to create 4 different combinations. Great starts and great finishes, great starts and poor finishes, poor starts and great finishes and finally poor starts and poor finishes. These are represented in a BCG(Boston Consultancy Group) Chart, which is my favourite graph. Many insights can be drawn from these tables.

First let us look at the great starts. In order to keep the tables to reasonable sizes I have to have cut-offs for each table.

1. Best career starts: over 125%

Batsman	Tests	Runs	Average	Start-Inns	NOs	Runs	Average	Ratio
RN Harvey	79	6149	48.42	15	4	1045	95.00	196.2%
TT Samaraweera	81	5462	48.77	10	3	581	83.00	170.2%
MA Taylor	104	7525	43.50	18	1	1088	64.00	147.1%
KD Walters	74	5357	48.26	16	3	903	69.46	143.9%
AJ Strauss	100	7037	40.91	20	2	1055	58.61	143.3%
N Kapil Dev	131	5248	31.05	14	2	510	42.50	136.9%
Javed Miandad	124	8832	52.57	17	4	917	70.54	134.2%
MEK Hussey	79	6235	51.53	18	4	957	68.36	132.7%
H Sutcliffe	54	4555	60.73	14	1	1037	79.77	131.3%
IT Botham	102	5200	33.55	12	1	479	43.55	129.8%
EdeC Weekes	48	4455	58.62	15	0	1125	75.00	127.9%
DL Haynes	116	7487	42.30	18	1	918	54.00	127.7%

Neil Harvey had the best start that any top cricketer has ever had. He had six hundreds in his first ten Tests and averaged 95.00. That is nearly 200% of his career figure. Thilan Samaraweera had a similar start, achieving over 170% of his career average. Mark Taylor, Doug Walters, Andrew Strauss and Michael Hussey should not surprise anyone. My hunch about Hussey was correct. But look at the starts Kapil Dev and Ian Botham had. They had averages like regular batsmen. Herbert Sutcliffe also had a near-80 average. It is surprising that only Sutcliffe, Javed Miandad and Hussey finished with a 50+ career average while the others dropped off.

2. Worst career starts: below 60%

Batsman	Tests	Runs	Average	Start-Inns	NOs	Runs	Average	Ratio
JH Kallis	162	13128	56.10	15	0	340	22.67	40.4%
SR Waugh	168	10927	51.06	16	3	271	20.85	40.8%
MS Atapattu	90	5502	39.02	19	0	321	16.89	43.3%
MD Crowe	77	5444	45.37	16	0	331	20.69	45.6%
HM Amla	70	5785	52.12	19	0	455	23.95	45.9%
HH Gibbs	90	6167	41.95	19	0	380	20.00	47.7%
DB Vengsarkar	116	6868	42.13	18	1	350	20.59	48.9%
ML Hayden	103	8626	50.74	16	0	413	25.81	50.9%
GS Sobers	93	8032	57.78	17	3	419	29.93	51.8%
DC Boon	107	7422	43.66	18	0	431	23.94	54.8%
DL Vettori	112	4516	30.11	17	4	230	17.69	58.8%
JL Langer	105	7696	45.27	15	0	402	26.80	59.2%

This is the other end of the table. Players, who had miserable starts to their careers. I knew about Steve Waugh's very poor start to his Test career, averaging 20.8. However, I could have never imagined that Jacques Kallis, who is currently averaging 56.1, started his career with an average of 22.67, which is around 40%, the same as Waugh's. Marvan Atapattu's starting sequence of 0 0 0 1 0 0 meant that he was going to be a contender for the worst start. It is a miracle that he has been upstaged by Kallis and Waugh. A knock of 149 runs in the 10th Test took care of that. But his average of 16.89 is the lowest by any batsman in this group at the end of the 10th test. It is surprising to see very poor starts by Martin Crowe and Hashim Amla, who currently averages over 52. But here comes the ball-to-Mike Gatting. Look at the awful start of Garry Sobers. A Kris Srikkanth-like 29.93. The amazing feature of this table is that four of these batsmen recovered very well, to post career averages exceeding 50.

3. Best career finishes: over 125%

Batsman	Tests	Runs	Average	Finish-Inns	NOs	Runs	Average	Ratio
S Chanderpaul	148	10830	51.82	15	4	1006	91.45	176.5%
SR Waugh	168	10927	51.06	14	4	863	86.30	169.0%
G Kirsten	101	7289	45.27	18	3	1003	66.87	147.7%
MS Dhoni	77	4209	39.71	15	3	700	58.33	146.9%
MJ Clarke	92	7275	52.34	18	2	1227	76.69	146.5%
CH Gayle	97	6836	42.46	17	3	859	61.36	144.5%
CL Hooper	102	5762	36.47	15	0	742	49.47	135.6%
KC Sangakkara	117	10486	56.99	19	4	1125	75.00	131.6%
DL Haynes	116	7487	42.30	17	4	719	55.31	130.8%
N Kapil Dev	131	5248	31.05	9	2	283	40.43	130.2%
RC Fredericks	59	4334	42.49	20	2	974	54.11	127.3%

These are the great finishers. They retired at the top, so to say. Shivnarine Chanderpaul, MS Dhoni, Michael Clarke, Kumar Sangakkara and Chris Gayle are still active and this value represents what they did in their last ten Tests. The amazing thing is the performance of Steve Waugh. Starting at 40%, he finished at 169%. Gary Kirsten, Carl Hooper, Desmond Haynes and Roy Fredericks all finished right at the top. Kapil Dev who appeared in the first Table has also finished well, though as a batsman. When I do the bowling article he is likely to find himself at the other end of the spectrum. It is safe to say that this finish helped Steve Waugh have a very good average.

4. Worst career finishes: below 60%

Batsman	Tests	Runs	Average	Finish-Inns	NOs	Runs	Average	Ratio
IA Healy	119	4356	27.40	17	0	138	8.12	29.6%
PD Collingwood	68	4260	40.57	13	0	202	15.54	38.3%
Mudassar Nazar	76	4114	38.09	16	0	249	15.56	40.9%
AI Kallicharran	66	4399	44.43	14	1	253	19.46	43.8%
SR Tendulkar	198	15837	53.87	16	1	367	24.47	45.4%
GA Gooch	118	8900	42.58	19	0	397	20.89	49.1%
TT Samaraweera	81	5462	48.77	18	0	440	24.44	50.1%
KJ Hughes	70	4415	37.42	19	0	372	19.58	52.3%
DR Martyn	67	4406	46.38	18	2	392	24.50	52.8%
MC Cowdrey	114	7624	44.07	17	0	396	23.29	52.9%
ST Jayasuriya	110	6973	40.07	18	0	393	21.83	54.5%
DB Vengsarkar	116	6868	42.13	16	0	370	23.12	54.9%
MP Vaughan	82	5719	41.44	17	0	388	22.82	55.1%
Inzamam-ul-Haq	120	8830	49.61	18	2	439	27.44	55.3%
ML Hayden	103	8626	50.74	18	1	486	28.59	56.3%
L Hutton	79	6971	56.67	16	0	511	31.94	56.4%
GR Viswanath	91	6080	41.93	15	1	334	23.86	56.9%
V Sehwag	104	8586	49.34	17	0	498	29.29	59.4%
H Sutcliffe	54	4555	60.73	14	1	471	36.23	59.7%

These batsmen finished very poorly. Ian Healy can be given a miss. Paul Collingwood, Alvin Kallicharran and Mudassar Nazar had nightmare finishes to their careers. But not at the same level as the next entry - Tendulkar. As already mentioned, Tendulkar, in his last 10 Tests, all at home, averaged 24.46. The writing on the wall is big and bright but is unfortunately not seen by many. "He should take the call" is the refrain used by all people, including even the latest entrants to the IPL gravy train. Would any other batsman have survived this level of performance? One really good innings, the fluent and very valuable-81 at Chennai, out of the 16 during the recent past. Graham Gooch also has finished below 50% too. Then a set of quality players have all finished below 60%. The fact that there are 19 players in this sub-60 list indicates that many batsmen stay beyond their sell-by date. And the presence of greats like Tendulkar, Len Hutton, Sutcliffe, Gundappa Viswanath, and Inzamam-ul-Haq in this lot is a matter to ponder over.

Now for the combination analysis. I have considered the two percentage values, start and finish and got them grouped into 4 combination groups. Not all batsmen are covered though. For each combination I have set different criteria and grouped the players. This is to ensure that the graph is not too crowded. The absence of players from these four groups basically means that they are in the circle in the centre. You can have a look at all players by perusing the downloadable file.

To represent these selected players I have used my favourite BCG chart. BCG stands for Boston Consulting Group for whom Bruce Henderson invented this method of depicting growth-share matrix for companies. It has since found many uses and I have used this graph extensively. When there are two independent measures, the quadrant-based representation of BCG chart is an excellent visual method of classifying players. In this case the two independent measures are the first-10-test % and the last-10-test %. These two measures form four groups and lend themselves to an excellent BCG representation.

We have the first-10-tests average percentage in the X-axis and the last-10-tests average % in the Y-axis. The graph is split into four quadrants. The top-right quadrant represents great starts and great finishes. The top-left quadrant represents poor starts and great finishes. The bottom-left quadrant represents poor starts and poor finishes. The bottom-right quadrant represents great starts and poor finishes. Let us now look at the graph.

Since I am going to show the tables for all these classifications I am not going to delve too long on the graph now. Let me highlight a few top players who are presented in the graph. Barring Kapil Dev, the top-right quadrant features many top players including Sunil Gavaskar, Chanderpaul and David Gower. Sutcliffe, Everton Weekes and Miandad are in the bottom-right quadrant. A number of top batsmen including Gooch, Michael Vaughan, Inzamam and Matthew Hayden are in the bottom-left quadrant. The top-left quadrant features Waugh, Sangakkara, Ian Chappell and of course, Kallis.

Now let us move on to the tables containing these group combinations. More players are featured in these tables than the graph.

5. Great start and great finish

Batsman	Tests	Runs	Average	START-Inns	NOs	Runs	Average	% of Career avge	FINISH-Inns	NOs	Runs	Average	% of Career avge
S Chanderpaul	148	10830	51.82	14	4	618	61.80	119.3%	15	4	1006	91.45	176.5%
MA Taylor	104	7525	43.50	18	1	1088	64.00	147.1%	19	2	905	53.24	122.4%
N Kapil Dev	131	5248	31.05	14	2	510	42.50	136.9%	9	2	283	40.43	130.2%
DL Haynes	116	7487	42.30	18	1	918	54.00	127.7%	17	4	719	55.31	130.8%
JH Edrich	77	5138	43.54	15	2	657	50.54	116.1%	17	2	801	53.40	122.6%
MEK Hussey	79	6235	51.53	18	4	957	68.36	132.7%	18	3	786	52.40	101.7%
SM Gavaskar	125	10122	51.12	20	4	978	61.12	119.6%	13	0	755	58.08	113.6%
DI Gower	117	8231	44.25	16	1	763	50.87	114.9%	19	4	776	51.73	116.9%
AB de Villiers	85	6364	50.51	17	1	841	52.56	104.1%	17	2	907	60.47	119.7%
CH Lloyd	110	7515	46.68	18	3	795	53.00	113.5%	14	2	611	50.92	109.1%
M Azharuddin	99	6215	45.04	16	2	711	50.79	112.8%	18	2	739	46.19	102.6%
AN Cook	90	7307	49.04	18	2	815	50.94	103.9%	19	1	947	52.61	107.3%

This table gives the full details of the combination analysis. The criteria here are that both percentage values should be over 100. Chanderpaul is still active. However a start of 119.3% and a finish (on-going) of 176.5% is magnificent. The player that Chanderpaul is, he is unlikely to suffer a drastic loss of form. Mark Taylor had an equally spectacular start and finish to his career. Kapil Dev is the only non-batsman in this list. Hussey had a wonderful start and a very good finish. Gavaskar, the true professional, knew when to quit, complementing his excellent start. AB de Villiers and Alastair Cook are currently active players who could go off this list if their current form drops. Let us not forget that Chanderpaul, Hussey, Gavaskar and de Villiers have achieved this with career averages of 50+. Chanderpaul averages 48.96 during the middle 128 Tests.

6. Poor start and poor finish

Batsman	Tests	Runs	Average	START-Inns	NOs	Runs	Average	% of Career avge	FINISH-Inns	NOs	Runs	Average	% of Career avge
VVS Laxman	134	8781	45.97	16	2	405	28.93	62.9%	19	2	569	33.47	72.8%
RB Kanhai	79	6227	47.53	19	2	505	29.71	62.5%	16	2	479	34.21	72.0%
MW Gatting	79	4409	35.56	18	1	390	22.94	64.5%	19	0	425	22.37	62.9%
Mohammad Yousuf	90	7530	52.29	18	1	588	34.59	66.1%	20	0	636	31.80	60.8%
DC Boon	107	7422	43.66	18	0	431	23.94	54.8%	15	0	463	30.87	70.7%
KJ Hughes	70	4415	37.42	19	0	513	27.00	72.2%	19	0	372	19.58	52.3%
MP Vaughan	82	5719	41.44	16	0	439	27.44	66.2%	17	0	388	22.82	55.1%
Inzamam-ul-Haq	120	8830	49.61	17	2	466	31.07	62.6%	18	2	439	27.44	55.3%
GA Gooch	118	8900	42.58	17	2	414	27.60	64.8%	19	0	397	20.89	49.1%
HH Gibbs	90	6167	41.95	19	0	380	20.00	47.7%	17	1	439	27.44	65.4%
ML Hayden	103	8626	50.74	16	0	413	25.81	50.9%	18	1	486	28.59	56.3%
DB Vengsarkar	116	6868	42.13	18	1	350	20.59	48.9%	16	0	370	23.12	54.9%
IA Healy	119	4356	27.40	14	0	231	16.50	60.2%	17	0	138	8.12	29.6%

This is the other end of the table. Batsmen whose starts and finishes were below par. The criterion is that both numbers should be below 75%. There are a number of batsmen in this list. Laxman's drop in form and indifferent start are fresh in our memory. Spare a thought for top class batsmen like Gooch, Hayden and Dilip Vengsarkar who had poor starts and finishes. This double low figures also indicate that the career of these batsmen between the 11th and 11th-last Test has been very good: much better than their career averages. Let me take the striking examples of Hayden and Gooch. Hayden averaged 56.40 during the 83 matches in between and Gooch, 46.22 during the intermediate 98 Tests.

7. Great start and poor finish

Batsman	Tests	Runs	Average	START-Inns	NOs	Runs	Average	% of Career avge	FINISH-Inns	NOs	Runs	Average	% of Career avge
TT Samaraweera	81	5462	48.77	10	3	581	83.00	170.2%	18	0	440	24.44	50.1%
H Sutcliffe	54	4555	60.73	14	1	1037	79.77	131.3%	14	1	471	36.23	59.7%
IT Botham	102	5200	33.55	12	1	479	43.55	129.8%	15	2	271	20.85	62.1%
Javed Miandad	124	8832	52.57	17	4	917	70.54	134.2%	17	1	569	35.56	67.6%
AI Kallicharran	66	4399	44.43	17	2	725	48.33	108.8%	14	1	253	19.46	43.8%
EdeC Weekes	48	4455	58.62	15	0	1125	75.00	127.9%	18	1	650	38.24	65.2%
MA Atherton	115	7728	37.70	19	0	840	44.21	117.3%	19	0	448	23.58	62.5%
V Sehwag	104	8586	49.34	13	0	693	53.31	108.0%	17	0	498	29.29	59.4%

These batsmen had a wonderful start but finished very poorly. The criteria are that they should have started at over 105% and finished at below 75%. Not many qualify. Samaraweera had the biggest difference between the start and finish, a huge 120%. Sutcliffe, Miandad and Weekes had substantial drops. Virender Sehwag's recent travails are reflected in this table. He averages at only 60% of his career. His return back to the Test team is uncertain. More so since he is not the one to decide when he should quit. He is under the selectorial hammer. Kallicharran probably had the worst end to his career.

8. Poor start and great finish

Batsman	Tests	Runs	Average	START-Inns	NOs	Runs	Average	% of Career avge	FINISH-Inns	NOs	Runs	Average	% of Career avge
SR Waugh	168	10927	51.06	16	3	271	20.85	40.8%	14	4	863	86.30	169.0%
G Kirsten	101	7289	45.27	18	0	568	31.56	69.7%	18	3	1003	66.87	147.7%
HM Amla	70	5785	52.12	19	0	455	23.95	45.9%	17	1	1010	63.12	121.1%
RC Fredericks	59	4334	42.49	19	0	525	27.63	65.0%	20	2	974	54.11	127.3%
KC Sangakkara	117	10486	56.99	17	1	638	39.88	70.0%	19	4	1125	75.00	131.6%
JG Wright	82	5334	37.83	18	0	438	24.33	64.3%	20	1	854	44.95	118.8%
IM Chappell	75	5345	42.42	16	1	409	27.27	64.3%	19	3	799	49.94	117.7%

These batsmen had a nightmare start and a fairy-tale finish. Steve Waugh: what can one say! He has one of the worst starts any player would have had, averaging just over 20 and finishes with an average of 86.3. This is the perfect example of quitting at the top. Generally Indians talk of Gavaskar's timing of his departure. But he averaged only 58 in the last 10 Tests. Look at Steve Waugh. Amla and Sangakkara are active players and their figures are bound to change. Look at the way Gary Kirsten and Ian Chappell ended their careers. At least Chappell is here because of his low career average. But Kirsten averaged almost 67 at the end.

Some readers might query that quite a few top players have not found a mention in this analysis. The point is that this article is about exceptions, on either side and in combination. Quite a few players are somewhere in the middle. Just to clarify this further I have given below the values for 7 top players across the ages who have not been referred to. This will clarify why these have not found a mention.

Hobbs        61  5410  56.95  785  46.18  81.1%  754  41.89  73.6%
Bradman      52  6996  99.94 1446  96.40  96.5% 1223 111.18 111.2%
Barrington   82  6806  58.67  727  51.93  88.5%  722  65.64 111.9%
Richards    121  8540  50.24  471  31.40  62.5%  550  39.29  78.2%
Lara        131 11953  52.89  812  47.76  90.3%  749  41.61  78.7%
Ponting     168 13378  51.85  670  41.88  80.8%  722  45.12  87.0%
Dravid      164 13288  52.31  773  48.31  92.3%  835  46.39  88.7%
Jayawardene 138 10806  49.57  750  46.88  94.6%  720  40.00  80.7%

Don Bradman just about missed the Great-Great combination. Richards just about missed the Poor-Poor combination. The others are right in the middle.

To download/view the list of 115 batsmen and the complete tables ordered in different forms, in Text file format, please CLICK HERE and to download/view the list of 115 batsmen and the complete tables ordered in different forms, in Excel format, please CLICK HERE.

The bowlers will be covered in Part-2. For bowlers there is no problem at all. The measure used will be wickets because 30 wickets at a bowling average of 25 is decidedly superior to 20 wickets at 20. It is essential for the bowlers to capture wickets. The comparisons should end there.