Rather than attempt a review of my subjective impressions of the season, what I'll attempt is some sort of objective analysis of why we finished where we did.  Although I would first like to quote from my first match report of the season just to show I do sometimes know what I'm looking at:  "David Lemi sprinted back into the Bristol half to take the ball.  He gathered it and ducked under a clumsy challenge from Fa'amatuainu.  He them sped away on a mazy run that left a host of embarrassed defenders trailing in his wake.  For a moment it looked as if the defence would close him down but he chipped over the head of Abendanon, gathered the ball and just about got over the line as James attempted a last ditch tackle.  Even as we witnessed this amazing piece of individual skill and opportunism people were marking this down as try of the season!".

What I've decided to do is compare Bristol with the team finishing top of the league and Worcester who finished in eleventh.  Just to remind you of the true horror:

The thing that is most obvious is that we won just two league games in the season and fundamentally this is the reason we were relegated.  But let's see if we can't be a little more sophisticated.  The thing that next stands out is the number of losing bonus points we picked up: 6.  Only three other teams got more and we only got 3 in the '06 - '07 season when we finished third.  Clearly this indicates that we narrowly lost a number of games that if we had won would have kept us in the Premiership.  So why did we lose those games?  Thinking back I can think of a couple when refereeing decisions cost us dearly and others in which we failed to hold on to lead as the game drew to a close, but are these impressions borne out by the statistics?

I've used information from Opta for much of this analysis and also my own records.  Consequently you may well notice some discrepancies between different data. In particular I'm fairly sure I've failed to note down which Arscott did what and may have double counted, but I don't think this is a significant factor in my conclusions.

The first thing I've looked at is the try scoring situation:

This table shows that Worcester were even worse at try scoring than us (although Wasps and we tied for second to last) but that their split between forwards and backs was much closer to the GP norm than either Leicester or ourselves.  The most interesting thing is the tries scored by the wings.  Both Leicester and Worcester score a below average number from this source, but Bristol got nearly half their tries from the wings.  Clearly David Lemi was a factor here, but Lee Robinson weighed in with 7.  This is a disproportionate source of tries for a side.  It might make some worry about our success next season with Lemi's departure, but I prefer to think that Mat Turner is a perfectly good replacement.  What it really makes me think is that if the 15 tries from the wings was a more normal 30% of the total then we ought to have been able to score around 50 tries in the season.  This would have had a significant impact on those narrow games and consequently our league position.

If we look at how tries were scored against our three teams:

Note: it is possible we actually conceded 77 tries as I've actually got two different figures in my notes, oops.

It is obvious that Leicester conceded far fewer tries and that Bristol were by far the worst in the league (contrasting with the 37 we conceded in our third place season).  Interestingly we conceded the majority in the second half and the vast majority of those in last quarter.  Perhaps this is evidence of problems with fitness.  There doesn't appear to be much of a difference here in the different positions except to say that our defence against wingers is good but we appear vulnerable through the centre.

Well, let's have a look at kicking.  First I have to point out that some of these data are a bit shaky.  Some of the kickers have so few attempts (step forward Greg Barden!) that it isn't really fair to compare them and also the figures for Hickey and Walker are hard to breakdown because so many sources fail to differentiate between their points scoring at their other clubs during the season.  But we'll see what we can come up with.  The first thing to note is the strike rate with conversions, where both Barnes and Walker are very similar at around 68%.  Leicester's main pair of Flood and Dupuy come in at 78 and 83% respectively.  Barnes' effort is actually quite commendable if we recall our disproportionate number of tries scored by wingers.  Jarvis and Tarscott have impressive 100% records!

On the other hand, if we look at penalties, Barnes hits about 57% and Jarvis is worse at 50%.  This does not compare well with Walker on 69%, Flood on 68% or Dupuy on 72%.  The subjective impression that Bristol's place kicking game is sub-par does appear to be reflected in the statistics.

The strangest thing about our place kicking was rather unexpected.  That is, the remarkably low number of penalties we actually attempted to kick.  Does it reflect a low number of penalties awarded?  I can't actually find published figures, but if so it might indicate a failure to put teams under pressure.  Of course, it does mean that if we don't attempt many kicks and our strike rate is low then we simply won't get as many points as other teams.

Moving on to our set-piece performance, we can see little difference between the teams in percentage terms.  The absolute numbers do show some trends with both Worcester and Bristol make noticeably more tackles than Leicester and this means that the number of missed tackles is greater despite the similar percentages.  The more missed tackle, the more points conceded.

We also see that Bristol had the second lowest number of rucks in the league which is perhaps another indicator of a lack of pressure on other sides and, possibly, problems with our support play.  We also had the largest number of scrums in the league, and unsurprisingly, the largest number lost.  I don't believe the degree of variation is great enough between teams for this to be a significant area.

In other areas of the game we do see a much clearer pattern.  We made the lowest number of passes (2302, 68% of the top team).  We had the second lowest offloads (63% of the top), the second lowest carries and the least clean breaks (61 against 159 for Leicester).  It is hard to avoid the conclusion that this is reflected in our try scoring performance.

What can we conclude from all this?  The top four teams in the Premiership were the four highest try scorers.  Even if you, unfairly, discount Leicester's tries against us in the last game of the season they still finish in the top four.  In contrast the bottom four consist of teams from the lowest try scoring five.  The team that bucks the trend is Wasps, who scored the same as us.  But of course you should remember that they were in the relegation fight until the last stage of the season, only pulling away late on.  Furthermore, Bristol conceded the most tries (even if you ignore the hammering at Leicester), followed some way behind by Worcester on 64.  Interestingly, the third most tries conceded was B**h with 48.  This shows two things.  Firstly that the two bottom teams conceded far more tries than those above them, amounting to over a quarter of the tries scored in the league. Secondly, that the benefit of scoring tries is disproportionately greater than the handicap of conceding them.

Fundamentally, this all goes to show that the single best predictor of league position is the number of tries scored.  Bristol scored too few and conceded too many.  Our nearest rival was the most similar to this pattern but gave away a lot fewer whilst scoring nearly as many.  It is a simple conclusion that our attacking play was ineffective.  The lack of tries was compounded by our lack of points from place kicks, in particular penalties.  Here our strike rate is low and we don't even attempt to kick many.  The factors that contribute to this pattern are illustrated by the low number of passes and offloads which are related to our try scoring ability.  The low number of rucks probably shows problems with our support play, also contributing to the low passing and offloading numbers.  The number of tries conceded, particularly in the later stages of the game are doubtless related to fitness issues which may well also impact on our performance at the ruck.

We need to score more tries and not just from the wingers.  We need to defend better.  Our place kicking must improve.  Underlying everything; we need to be fitter.  We already knew this, I think, but it's nice to see it backed up with figures!

View a Printer Friendly version of this Story.

Bookmark or share this story with:

• Digg
• Delicious
• Facebook
• Reddit
• Stumbleupon