QUOTE(teoani @ Mar 2 2009, 02:45 AM)

Tom, being able to sight-read such complex pieces is really impressive!
It is not that big a deal. Rule of thumb is that you should be able to sight read 2 grades lower than your playing standard. My sight reading is not exceptional. It is about what you'd expect for a pianist of my standard, and probably inferior to many of the more modest contributors to this forum.
QUOTE(maledictis @ Mar 2 2009, 11:57 AM)

QUOTE(teoani @ Mar 2 2009, 02:45 AM)

I am wondering if consistent attempts at playing pieces at first sight can help me improve. I have been taking some random pieces from old books and murdering them by playing through slowly (with mistakes). Not sure if I am going in the right direction though. Currently the difference I feel is that I can look at the piece given during the sight-reading component of the exam, and think,"Ah, this doesn't look as scary as the one I played last night. Maybe I can manage" But the increased confidence hasn't been translated into increased accuracy yet.
It just takes time - a lot of time. I can sight read gr8 and above if they're not too "modern" (i.e. random timing etc.), but this comes from years of being a ballet pianist and accompanist and just having to play whatever was put in front of me for 3 or 4 hours a day - you sure learn to sight read quickly under that sort of regime...
It is good point about not being "too modern". Familiarity with the style matters. I can sight read a lot of Bach, Scarlatti, Haydn, Mozart, Beethoven and their contemporaries tolerably well - including many pieces that are set for diplomas up to fellowship level. But outside the classical/early-romantic period I am all at sea and the more modern the piece the more I have to put in a lot of work to make anything of it.
For example, from this year's Grade 8 list I can play reasonably at sight much of list A and almost all of list B but not a lot of list C.
For overall improvement in pianistic skill I haven't found that sight reading through lots of stuff helps me not much at all. It is useful for exploring repertoire and for acting as an accompanist, but not as a significant way of increasing your skill.
I suspect that what maledictis used to do involved a lot of repetition of the same pieces. That is certainly the case when you are playing for ballet classes, or for singers rehearsing. Only the first run through is sight reading. The rest is practice. But it is practice in a situation wher you are very motivated to get the essentials right!
For me what improves my playing more than anything else is learning a single piece in excrutiating detail. Which means when you think you know it, can play it from memory with ease, can safely negotiate the tricky technical corners, and your friends think is sounds good enough to be worth listening to THEN you start to really work on it! I don't know how universal this is, but it is true for me, so it will be true for at least some others, and possibly many or most others.
What is strange - at least when you first think about it - is that this process improves my ability to sight read more than actually practicing sight reading.
[Aside:
You might be interested to know that there are simlarities in learning to play better chess. There is a form of chess, speed chess, where each player is allocated a relatively short amount of time on their clock"- most popular is 5 minutes each. If the game is not finished already on the board then the player who runs out of time first loses. It is quite analogous to sight reading a piece of music. You have no time to think deeply.
It has been proven in several pieces of research that playing lots of this 5 minute chess has very little effect on a player's skill level. The most effective way to increase skill at chess is to study a single game, or even a single position from one game, in great depth so as to understand it in minute and intimate detail. The first time you do this exercise it might take weeks, or even months.
And the perverse effect is that this slow in-depth form of study improves your ability to play 5-minute chess more than any amount of playing 5-minute chess.
end Aside]
In actual practice the most important thing for me is to master difficulties by practicing extremely short sections. It takes great discipline to avoid the temptation to bash through pages and pages, but it is worth the effort. Progress is at least 10 times faster. To master a difficulty requires focus, intensity, and repetition. So by repeating the difficult skill many times, in a short period of time, the mental patterns seem to be quickly "burned in" to the brain. In contrast if you play a large section, with many tricky parts, then the intensity and focus just is not there. Of course it is necessary to play whole pieces to establish balance and continuity, but that is a relatively small part of practice, especially in the early stages.
The next most useful thing I have discovered is the value of recording yourself and comparing what you really sound like with what you thought you were doing. All the books tell us that you must listen to yourself as you practice, but it is easier said than done. That is because for most of us, especially with a new piece, so much of the attention goes into the physical movements that little is left to listen intently. When a piece is so well known that playing it has become effortless THEN you can really listen to what you are doing.
In the interim a recorder comes to the rescue. And reveals huge numbers of inaccuracies that you did not notice while you were playing. For me it was at first very disheartening to hear just how awful I sounded - with uneven basic tempo, random and unintended hesitations, speed-ups, slow downs, exaggerated dynamics, misplaced accents, notes unintentionally loud or soft, notes held beyond their proper value, mistimed pedalling ... you name it. But by repeatedly recording, listening back, re-recording very quick progress can be made to establishing better control. This is one of the ways that modern technology replaces what used to be one of the teacher's main functions.