| |
BGonline.org Forums
Rules of Thumb to avoid crunching?
Posted By: sebalotek In Response To: Rules of Thumb to avoid crunching? (Timothy Chow)
Date: Friday, 18 September 2015, at 9:35 a.m.
Thanks for everybody’s input so far – much appreciated.
New Question: it is inevitable that we will all lose games on occasion. Roughly what *proportion of game losses* should I realistically be aiming to lose while severely crunching? I’ve estimated my current crunch rate to be approx. 20%. What’s the approximate average? 5%? 15%?
Casper –
Thanks for your good advice. Even though I play mostly by instinct (lack advanced maths), my opening moves are not too bad. On analysis, I generally seem to split when I should split and run when I should run. I rarely see dings for opening moves (sometimes with early doublets or at unfamiliar cube scores). But yes, I’ve never memorized even up to 3rd roll so although I might be making very small errors, perhaps these are not big enough to get flagged. I should aim to learn 3rd roll rote like. Maybe these small errors are becoming compounded and magnified with immediate unlucky (or opponent’s lucky) rolls. I really should check out http://backgammonstudio.tk/ for practice!
Great idea about the stories: I had already made a simple tally list categorizing the tipping points of how I lost each match, i.e. primed / crunch, move blunder, lost race, dropped take, wrong pay now, wrong pay later, anti-joker, CPU joker. I probably missed some categories.
Even so, this requires the huge effort of placing my notepad and pen next to my CPU and (more importantly) finding the discipline to always tick boxes after each game. I will see if I can lift that pen! :)
P.S. It would be revolutionary if the next generation of backgammon programs could perform this sort of auto-analysis and boring error categorization / insight generation for you. GNU / XG: are you developing this already? :)
Mike –
>> You should think about splitting your back checkers at the earliest convenient opportunity. Especially if your opponent strips their 8 point.
Good tip. I mostly split asap already unless I am trying to blitz or mini-prime – but perhaps I need to try it more. Regarding the stripped 8 point, I already do this but find I still get hit a lot after running to my opponents outfield. And frustratingly, even if I am missed next roll, my immediate roll after often suggests consolidating / safety-ing the investment in this runner which often leads to a stacked position on my 13, 8 or 6 point along with the associated inflexibility this brings. I don’t like it, but I guess there is no way to avoid this (fairly regular) post-run stacking?
Chase –
>> depending on the position and the dice
Yes, these variables mean that there is no ‘silver bullet’.
>> When you review your games, look for conceptual errors.
Great advice. Trouble (for me) is, this gathering of data takes valuable time plus some ‘effort’. Ultimately rewarding, but easier said than done. Solution is Nike’s tagline!
>> The challenge is to develop the judgement to know which is the right tool for the job at hand
There’s the rub! Especially when you have two or more rules of thumb which contradict each other for a given position. I often have trouble in these situations where both seem equally important.
>> The key is to improve your judgement about how the game is most likely to evolve and make moves that anticipate it. Constantly ask yourself, How are things going to look two or three rolls from now? (And constantly check to see if you were right, and, if not, why.)
I tried this last night, even occasionally vocalizing ‘How are things going to look two or three rolls from now?’ as I found I tend to slack and play ‘obvious’ moves quickly, on auto-pilot. I found that prediction does work sometimes, but quite often the randomness of the dice and explosion of permutations demands a wholesale change in strategy from even one move to the next. My feeling is that this randomness diminishes the value of this cognitive effort (and accuracy) of my attempted predictions. But I also feel that I should try and persevere with your tip.
Timothy –
>> you are too afraid of her attacking potential and not afraid enough of her priming potential.
Thanks – this is a possibility. I never thought of it in these terms before.
>> You can see that you might get hit and closed out.
Damn straight! :)
>> You have to see that several moves down the line your back checkers will get trapped.
..or that I will run, get hit, and *then* get trapped.. Argh!
>> split and/or run your back checkers more boldly in positions where there is some danger of getting attacked but even more danger of getting primed.
Great advice – I will actively try and integrate this.
P.S. Interesting how you anthropomorphized my CPU to be a female :) I have always vaguely thought of my computers as males. I tried, but don’t think I / they can change at this late stage.
Final thought: if the computer rarely crunches, even when I win, perhaps more aggressive running isn’t the answer. Perhaps I need to look for opportunities to develop my own prime oriented game – after all, it works for the CPU..
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.