The rule of thumb that an army would start to disintegrate after losing about one third of its strength applies during the shock and stress of battle. (This is particularly important as such disintegration is what causes many battles to be so decisive; two similar size armies could slog it out for hours, suffering gradual losses, but when one army finally breaks it usually lost a much larger number of men.)
And Sun Tzu's estimate would vary greatly depending on the quality of the army and its current coherence. Highly motivated armies could lose very little, while poorly motivated ones could lose more.
Forced marching was as likely to reduce the battle effectiveness of an army as it was to actually reduce the army's numbers. This would be through fatigue, stragglers or similar, all of which could be recovered with sufficient rest.
So, if the aims of the march could be achieved at the reduced effectiveness (say, capturing a key defensive point before the enemy), it may be possible for the forced march to have little long term impact.
An example of both the advantages and disadvantages of forced marching can be seen in the example of Harold Godwinson's campaign in England in 1066. Facing off against a possible invasion by the Normans along the south coast, he was informed of the invasion and capture of York by Harald Hardradr of Norway.
He marched his bodyguard (highly motivated & disciplined!) north and caught the Norse away from camp without their armour, and defeated them soundly. Harald Hardradr was killed with an arrow to the throat.
He then marched them south again, to find William (soon to become 'the Conqueror') ashore, and stood against him at Hastings. Harold's loss to the Normans is at least partly attributed to the fatigue of the two forced marches his men had had to endure.
A couple of notes here: Harold chose to march with only his best troops to maximise speed and minimise losses; and the value of surprise against the Norse outweighed any disadvantages from the march.