Monday, 6 February 2017

Staying alive

Sadly not a Steve Beaton post. In the wake of the Super Bowl, we have a situation that comes up a lot in darts that is similar to American football - you have a spot where you are deep in opposition territory down by, say, 10 points, and you are on fourth down with not long left. The conservative play is to kick the field goal and make it a one possession game, but it's usually the wrong play, as it takes time and skill to get all the way down the field to then score the game tying touchdown, when you've already done that - you need a touchdown, so go for it now, then you know what you need to do if you don't get it, if you do, you then only need to get into field goal range, which is somewhat easier. There's similar spots like, when losing by 15 points, teams will usually kick an extra point after scoring to ensure a 1 possession game, which then goes horribly wrong if they mess up a two point conversion later - go for the two now, and you have time to play faster knowing you still need two scores.

The darts comparison comes with checkouts of things like 132 or 135. The common play is to go for 25/bull first, as that keeps things alive assuming you hit. But is it maximising your chances of winning the leg? If you hit 25, you still need a treble to then leave a shot at bull - which if you miss, is likely to still leave you on a two dart finish, either 25 or missing completely and not lucking out and hitting single 10, 18 etc. The bull could be obscured by your first dart, but at the same time, it could easily leave you with a good marker so I'm inclined to say that may end up evens in the long run without any observed evidence to the contrary. If you miss your treble after hitting 25, you're on 88/90, which is still going to need a treble to leave a 1 dart outshot.

Hitting the bull first on either checkout leaves you with a treble-double outshot, and in both cases you still need a treble to leave a 1 dart outshot if you miss. Of course, you can go for a second bull on 132, which leaves fat 17 for tops if you hit 25.

The question is whether this is more likely to kill the leg than not going for bull first? Say you go for treble 20 on 132, or treble 19 on 135. If you hit, then in the first instance you can stay there for double 6, particularly if it's a good marker, or go double-double, either 18's twice or 20's and 16's. In the latter case, you can go double 19-double 20, try for treble 18 for double 12, it'll be situational depending on what your opponent is on. If you miss, then just stay on the same target and get four more of the same score to leave double 16 (you can deviate to other doubles if you hit treble second dart obviously) or tops.

Is this more likely to finish than going bull first? If you have enough time for two visits, it seems almost certain. If you only have enough time for one visit, then you're going to need to run some numbers as to how often you hit each target and work the probabilities from there. My gut feeling would be that going bull first kills less often in that visit, but people seeing that your chances of finishing if going for a treble is zero, rather than something larger if you hit some part of the bull, lead people to go for the worse route knowing they have a backup plan, regardless of how unlikely it is.

Your opponent's score is huge in this as well - going bull first if your opponent is not on a finish is idiocy, but what if he's on 161? This seems equally mad. If going bull first finishes more in one visit than not, then it needs to do so often enough that it outweighs the times you lose the leg by either your opponent finishing, which doesn't happen very often, or by your opponent finishing on a second leg because your bull usage has left you a two dart finish rather than a probable one dart finish at an easy double, which you then miss.

Stats for how often people finish on various outs already exist - anything beyond 120 is already, at best, around a 10% chance. The rest would need to be worked out, it's a case of getting the data. I'd assume that some players have already done the sort of analysis needed (probably the Dutch I'd guess), but you never know. Of course, the trick is to start counting earlier and hit the bull the visit before to not leave the shot in the first place.

No comments:

Post a comment