Declarer play comes with a built-in feedback loop. You choose a line, it works or it doesn't, and by trick thirteen you generally know which. Defense offers nothing so tidy. You follow suit, the hand rolls on, declarer claims, and the score goes in the column. Did your card cost a trick? Nobody at the table knows. Nobody is going to tell you.
That silence is exactly why defensive errors survive for decades. They are rarely dramatic. They are small, repeatable, and invisible from your own chair.
Leading on autopilot
Fourth from the longest and strongest is a reasonable default. A default is not a plan.
The auction almost always points somewhere. Declarer bids two suits and runs to 3NT. The opponents crawl into four hearts after a long, unhappy conversation. Partner overcalls and you hold three small cards in their suit. Those are three different problems, and the gap between a passive lead and an attacking one is frequently the entire board. Your opening lead is the only card you ever choose knowing the whole auction and nothing about dummy, which makes it both the most information-poor decision you make and the highest-leverage one. Opening leads that win boards digs into the choices themselves. The habit worth building is smaller than that: pause, ask what the bidding just told you, then lead.
Defending to the wrong target
Four spades. You can count two tricks in your own hand and probably one in partner's, so you start cashing.
At rubber bridge or IMPs that instinct is usually right. Take the sure tricks, beat the contract, bank the plus score. At matchpoints it can be a quiet catastrophe, because if the contract is cold and the whole field makes it, the only thing separating you from everyone else is whether declarer took ten tricks or eleven. Defending an unbeatable contract has nothing to do with setting it. It is about holding declarer to the minimum, and that often means not cashing, because every winner you take early may hand declarer a discard later.
Same cards, different scoring, completely different defense. Players who never make that switch leak a fraction of a matchpoint board after board, and none of it ever looks like a mistake.
Taking the ace because it is an ace
Declarer leads a small card toward dummy's king. Your ace is on the table before you finished deciding to play it.
The old rule about playing second hand low exists because that reflex is so strong and so often wrong. Rising with the ace on the first round gives declarer a free trick with the king later and announces exactly where the missing honour lives. Ducking smoothly costs nothing when declarer was always going to score that king, and it wins a trick outright when declarer holds a singleton or is simply guessing. Your ace is not going anywhere. Aces keep.
Signals that nobody reads
Partner leads the king of diamonds. You hold Q-7-3 and dutifully contribute the seven, or the three, or whichever card your agreement calls for. Partner then shifts to a spade anyway.
Signalling fails in two directions, and most partnerships manage both. Some players signal beautifully and never once look at what partner played. Others watch every card and signal from habit rather than from what they actually want to happen. Both halves have to function or the entire apparatus is decoration. The fix is unglamorous. Before you drop a spot card, decide what you want partner to do at the next trick, then play the card that says it.
Not counting, then blaming the layout
Most defensive disasters that get written off afterwards as bad luck were countable at the time.
Declarer showed five spades in the auction, dummy arrived with three, you hold two, so partner started with three. Dummy has nine high card points, you have eleven, the opening bid promised twelve to fourteen, and partner is close to broke. From there you can usually work out whether the finesse declarer needs is onside and whether your only real hope is partner holding one specific card. Counting is not a talent. It is a habit that feels impossible for about six sessions and then quietly becomes automatic. How defenders read declarer's hand walks through the mechanics.
The mistake underneath all the others
Every error above shares one accomplice: nothing tells you it happened. A declarer who butchers a hand at least sees the contract go down. A defender who butchers one watches a normal-looking contract make and moves on to the next board, entirely at peace.
Reviewing with partner afterwards helps, and it is worth doing, though two people who just defended badly make poor witnesses to their own defense. What settles the question is seeing what happened when other people held your cards. If three other Wests found the diamond shift and beat the contract while you cashed your ace and let it through, there is no argument left to have. That comparison is the entire reason duplicate scoring exists, and it is what Bridge@Home brings into a living room: your table of four plays pre-dealt hands that other players have also played, so afterwards you can go board by board and find out whose defense actually held up.
The takeaway
Defense is the half of bridge that nobody watches, yourself included. Slow down before the opening lead. Decide what you are defending for before you start cashing. Let the ace wait. Count the hand, badly at first. Then go and find out what everyone else did with your cards, because the only defensive mistakes you can ever fix are the ones you know you made.