Thursday, 28 July 2016

Darts wave fone-tuning [sic]

it may be a little bit early to talk about fine-tuning, or "fribbing", as I believe some people call it. Fribbing is apparently more fine than frobbing, just as twiddling is more fine than twoddling. So I guess you could say I'm not yet fine-tuning, but just fone-tuning.

Ask Steven Pinker for a clearer explanation of that first paragraph if it made no sense.

Anyway, some things I've noticed as a result of having used my darts wave a few times and made a few improvements (see the last few posts of the Massive Blog for more details) are:
  • that it remains very good, 
  • that it indeed keeps getting better as a result of the improvements
  • that it does help to use it a few times to find minor errors in the spreadsheet
  • that if you have a momentum score based on the last 15 visits to the board then this can go down when you throw a ton-40 if you threw a ton-80 16 throws ago.
This last point is quite interesting. Clearly, it doesn't matter how many throws you take into account. If the 180 leaves the relevant data range at the same moment as the 140 enters it, then your momentum will drop. Just as your ranking points will drop in tennis or whatever if you are only runner-up in a tournament which you won the year (or two years) before. This seems largely unavoidable.

However, there is one points-based system which I believe is universally revered and respected, and that is the FIFA world ranking system for international football teams. I don't want to get into the nitty-gritty so I'll stick to the notty-grotty, which includes the fact that if a football match was three-and-a-half years ago, then it contributes less to your ranking than if a match was last Wednesday.

I haven't experimented yet, but this has certainly given me the idea of making the last 5 visits to the board of my wave count more, the 6th to 10th visits count less and the 11th to 15th visits count even less. This could potentially counteract the issue of your ranking dropping when you hit a 140. Probably not in every case, but possibly in many cases.

Worth thinking about anyway. I will probably do a couple of trial runs and post something about them in the near future. In the mean time, why not read my other posts about the FIFA world rankings, like this one here.

PS: Just in case you're interested but not interested enough to have clicked on the link, the current top five world football teams, in order, are: Argentina, Belgium, Columbia, Germany and Chile.

Wednesday, 27 July 2016

Darts wave at the World Matchplay

I trialled my darts wave at the World Matchplay, which finished at the weekend, and I can report the following:
  •  it worked well 
  • it was possible to keep up-to-date with the flow of a match 
  • only when I sneezed or had an itchy foot or something did I miss a couple of darts
  • missing a couple of darts doesn't have too dramatic effect on the whole wave
  •  it was quite fun to do.
Here are a couple of the waves I made.


During the tournament, I also added a macro to my spreadsheet, which basically exported the graph as a JPEG with a unique name (numerically adding +1 to the file name each time). Using these many JPEGs I could then make GIFs of the wave, like this one for the final.

[Blogspot isn't letting me copy the GIF in here, but you can see it here.]

UPDATE (08.08.) here's another stab at entering the GIF:

This is quite fancy and nice.

It was particularly nice that it was so quick and easy – saving the JPEGs really didn't hinder the entry of the data at all – and the GIFs almost made themselves as it was always just "all the images in the folder". And because of the way the file names worked, they were automatically in the right order too. This meant that I could quite comfortably grab an apple and make a GIF everytime there was a break (every 5 legs).

One thing that I have realised, having trialled the wave, is that after a few legs the impact of any 180 or missed double is comparatively less than at the start, which means you have to look quite closely at the graph to see the upturns and downturns. This is fine – it's nice to have an overview of the whole match, but I have also had another great idea.

Based on the same data entered, I have now set up the spreadsheet to generate a second wave which only considers the results for the last 3 legs ish (technically the last 15 visits to the dartboard each). This gives you a much more clear up and down, like this is how it would look for the final.

I only thought of doing this after the final, so I don't have it as a GIF, but I've now also extended the macro so that it saves both waves, meaning that from now on I'll be able to generate both GIFs, one for the trend of the whole match and one for the rapidly chanding 3-leg (ish) momentum. This is pretty exciting. Even having the still images is quite nice but the animations are even more fun.

I have also made a couple of other adjustments to the spreadsheet (which I haven't yet actually put to use in a match situation), which are just the addition of a "break" parameter and a "bullseye/double (not checkout) parameter. Adding and removing parameters is fairly easy, though I haven't made it super clean just yet. In theory I could add 10 placeholder parameters which you could kind of turn off an on as you fancied. This is all fairly small potatoes compared to the 3-leg (ish) wave though and other such innovations like the GIFs.

On the whole, very happy with the results of the first tournament the wave has been used at, which have:
  • confirmed the general functionality of the wave
  • led to cool new features
  • revolutionised darts broadcasting.
See you next week.

Monday, 11 July 2016

Still banging on about my darts wave

here's the wave for MvG (green) against Barney (yellow, because I had it set to Chizzy and couldn't be bothered changing it) from their excellent match at this year's World Championships.

This one.

One thing I realised when putting this together is that you should really get some momentum for an intentionally hit centre bull, even when it's not a bull finish. Maybe the "big treble" should be big treble or centre bull. It isn't all that important but worth thinking about.

Another quite important thing on this is that I upped the parameter for a double. It was at 2 and missed double was at -1.5, which meant that if you missed two doubles and hit the third you ended up being a bit too negative. A good example that, though, of somewhere to fine tune. In the example I upped it to 3, but you could up it more. This would be how the wave would look if a hit double was 5 points.

Might not look hugely different, but the interest is in the detail, as they say.

Enjoy anyway.

PS: I've deliberately not provided any analysis but one fairly obvious thing is that MvG basically moves onward and upward, albeit not always with the same constancy, and Barney basically nips in with a little push upwards to take his sets. I'll leave it up to you to decide whether or not that's interesting.