Risk taking should almost never be necessary! Just try to play the best moves!

Antillian wrote on 01/30/10 at 13:05:02:
This is the usual bad advice from chess players to chess players. The kernel of the problem is described very well here: http://www.chesscafe.com/text/heisman06.pdf Too much speculation too less knowing in the answer given here.

For these kind of questions there is no general answer. A good answer depends on playing level and profile of the questioner. (Btw: It is addressed in the article and in the title of Silman's book. Before re-assessing something you have to assess something. Silman's book is no book for beginners till average level, which is ~1600.)

Even super GMs routinely make less than perfect moves (mistakes) during a game.  If you don't recognizing your opponents mistakes then why would you expect to win?  Just play on and when you finally see a mistake, pounce on it, otherwise try not to lose.
And one other thing, a desperato attack when you are losing is not really risk taking.       

Taking pawns and pieces usually works better

You should take risks, but only when they are rational and calculated. However, the intended 'risky' move does not always have to be objectively the best - sometimes a speculative sacrifice that complicates the position significantly, creating a dangerous and sudden attack against the opponent's king where the opponent must calculate accurately, can decide the game in your favour. 

I advise against taking irrational risks, such as sacrificing material for minimal compensation (except in positions where other moves lose more quickly or easily), playing for a win in a drawn position but only creating losing chances for yourself, and seriously weakening your own position without achieving any compensation in return.

The philosophical way of viewing risk-taking in chess is that every move is a risk of some sort since the position changes with every move, and squares/pieces are weakened with every move, but obviously this isn't of much help to the player except for recognising what one can currently play, which could not be played on the previous move. 

Generally speaking, in an equal position against a slightly weaker player, you need to take some slight risks to play for a win, but if you do so at the expense of playing good moves, you are setting yourself up for disappointment.

In team competitions, a good general strategy (if you are the stronger team) is to play with the draw 'in hand' whilst playing for a win, and being ready to take risks if your team needs you to win. If you are the weaker team, offer a draw out of the opening, and then be ready to complicate the position without taking too many undue risks, and make sure you don't fall far behind on the clock.

If your opponent don't make blunders, and you are of equal strength, do your always need to take risk in order to win? E.g. make a sacrifice or open lines to your King.

If you take a risk, do always see how the game will continue?

How certain are you that your risk taking will get you an advantage?