In the first article (here) two questions turned out to be important when considering 'penalty double or pass': can the opponents escape to a better contract and, if yes, will they?
Sometimes another issue influences our decision to double or not: often contracts are only beaten on a special opening lead. If the player contemplating the double is not to lead himself he must ask himself what influence his double may have on partner's choice of opening lead.
If he fails to do so, he can regret that very much, as is shown in the following comical (to NS, that is) catastrophe. The year is 1964, the place is Toronto, the tournament the Men's Pairs (MP-scoring) during the ACBL Summer Nationals.

The name of the leading figure, East, was not mentioned in the account of this deal. Unfortunately that was and still is the habit in bridge-reporting: a player who shines gets full credit, a blundering player stays nameless. The reporter, usually a top class player himself (or ex-top class player), is obviously afraid of arousing the wrath of the player in question. Thus bridge-reporters humour each other: unprofessional, cowardly and bad for bridge as a top sport. Have you ever read in a football (soccer!) report: 'A Manchester United striker missed before the goal'...?

Anyway, the unnamed East got:

N/EW
				♠	A K Q 7 5 4 3
				♥	—
				♦	9 4
				♣	6 4 3 2

¹ Conventional, game forcing
² Blackwood
³ South: 'We have no agreement about my partner's 6♥ bid'

It's obvious NS have a severe misunderstanding: it is safe to assume South's interpretation of 6♥ differs from North's.
East doubled licking his lips, expecting seven undertricks, +1300 in those days, at present that would be +1700.
South had the nerve to redouble ('now we're talking +2600', East thought) and...

N/EW		♠	—
		♥	A K Q J 9 7 6 4 2
		♦	7
		♣	K Q 5
♠	10 8 6 2			♠	A K Q 7 5 4 3
♥	10 3			♥	—
♦	5 3			♦	9 4
♣	J 10 9 8 7			♣	6 4 3 2
		♠	J 9
		♥	8 5
		♦	A K Q J 10 8 6 2
		♣	A

...after a long and agonising pause West led the ♣J!
East was flabbergasted and shortly afterwards had to chalk up −2930 (nowadays that would have been −2980) instead of the expected +2600. West's failure to lead a spade made a difference of 5530 points.
It is not known whether South did claim seven redoubled overtricks ('+4330'). Some jokers do; after all, following this lead declarer has in principle 20 tricks...

Of course it was East's own fault. Presumably West was right in not leading a spade. A double of a slam is in principle − unless the slam is bid as a save − a Lightner-double. The doubler requests his partner to lead an unexpected (in view of the bidding) suit. The doubler thinks the slam will only be beaten on that lead.
If a voluntarily bid slam goes down, it will usually (!) not be defeated by many tricks. The goal of such a double is therefore not to obtain a higher plus score but to score positive at all. East could double here for instance if he were to hold something like eight spades to the ♠QJ and the ace in another suit, fearing NS to make 13 top tricks on a spade lead (which suit West is to lead then, is another thing; after all, NS haven't bid a single real suit...).

One thing is certain: if East had passed, West would have led a spade!
If the given auction happens at IMP-scoring, East should pass. With some regret of course: 'A pity we're going to score only +350 instead of +1700. But doubling is not an option, since it tells partner not to lead a spade.'
At the actual MP-scoring East's double was even more incredible: did he really think +350 would not have given him a top...?
To be completely honest: it wouldn't have! Another NS-pair ended up in 7NT doubled. At that table East's double was perfectly reasonable... he was to lead himself!
Anyway, 'our' East should have passed, scoring second best. A lot better than doubling and scoring a bottom...

We move to Ocho Rios, Jamaica, the final of the 1987 World Championship teams (contesting the Bermuda Bowl), USA − Great-Britain. Once more we see the Americans Martel-Stansby in action (they played an important part in the two deals of the first article):

E/—
				♠	A 10 8 4
				♥	K Q J 10 5
				♦	A
				♣	10 7 4

¹ Flannery: 11-15 HCP, 4-5 in spades-hearts

Decide for yourself first: should East double or not?

The advantage of the double is that, if NS stay put, EW can score +300 (after a heart lead East is certain to take six tricks: down two) instead of +100 (undoubled), a handsome extra 200 points.
Stansby thought this reason enough to double. Oops:

¹ Stansby did not double again; as a cool professional he doesn't do spite-doubling, since this costs too much, too often

Martel led the ♠2:

E/—		♠	Q 9 6 3
		♥	8
		♦	K 10 9 8 7 4
		♣	A K
♠	J 5 2			♠	A 10 8 4
♥	9 6 4			♥	K Q J 10 5
♦	Q 6 5 3 2			♦	A
♣	8 5			♣	10 7 4
		♠	K 7
		♥	A 7 3 2
		♦	J
		♣	Q J 9 6 3 2

It is clear that the double turned out wrong. Firstly because NS could — and did! — escape. South, Sheehan, felt that 3NT was not the right contract and fled. After all, he was somewhat weak for his 3♣ overcall (which shows, at the three level, the strength of a good opening bid) and mostly: he missed the two top cards in his long suit.
Secondly: although the escape contract (5♣) could have been beaten by two tricks as well, it turned out harder to do so. Had East passed 3NT, defeating that contract would have been a piece of cake.
The conclusion must be that East should have stuck to the 'bird-in-the-hand'-principle.
By the way: according to some experts the double is dubious for one more reason. Since this contract will not go down a lot usually and since East has shown two suits, the double may be looked upon as a Lightner-double, asking for a special lead, not the normal one. If so, East will pass if he wants the (normal) lead in his longer suit (hearts). By doubling he would ask a 'special' lead, in his shorter suit (spades). So East will double with something like:

E/—
				♠	K Q J 10
				♥	A 7 5 3 2
				♦	A
				♣	10 7 4

On the other hand: it is unlikely that this top level EW pair had not agreed upon the meaning of the double. One would therefore expect Martel to have led a heart against 3NT doubled (still: against 5♣ he led a spade, as we have seen...).

At the other table the Englishman Forrester (East) was left to play 2♥ peacefully. He went down one: -50 and an 8 IMP profit to Great-Britain.

The last deal is from a Bridge-Olympiad match between England and Ireland (IMP-scoring). The Irish (EW) bidding system was very uncommon (to say the least) and it created a bizarre problem for Englishman Kirby (South) :

W/All
		♠	—
		♥	A K Q 10 6 5 2
		♦	A J 8 3
		♣	9 6

¹ One unknown major and one unknown minor suit, 10-12 HCP

Obviously East did not correct to 2♠, since he judged from his own hand that West's major suit was hearts (hardly surprising in view of South's heart suit!).
If South were to double now, that would of course have been for take-out, not penalties. Probably North would keep on bidding spades (unless West was there first of course...). Defending 2♥ doubled was clearly not possible.
Bidding hearts himself wouldn't help South either, since by bidding 'the opponents' (!) suit' he would not show hearts (having said that: if South were to double, it was to be expected that West would escape, after which a heart bid by South perhaps would be natural after all ...).
Anyway, it looked (too?) difficult for NS to end up in 4♥. 3NT was not an option either, since South lacked guards in both black suits!
So Kirby decided to settle for three or four undoubled vulnerable undertricks: he passed.
Alas, instead of +300/+400 he scored +700:

W/All		♠	Q 6 5 3
		♥	8 7
		♦	K Q 6
		♣	A J 8 7
♠	A K 9 8 4			♠	J 10 7 2
♥	J 4			♥	9 3
♦	10			♦	9 7 5 4 2
♣	K 10 5 4 3			♣	Q 2
		♠	—
		♥	A K Q 10 6 5 2
		♦	A J 8 3
		♣	9 6

By the way, England won an IMP, since the Irish at the other table missed the slam:

It turned out 6♥ wasn't that easy to bid.
Perhaps North might have given a little push. After all, South had jumped to 4♥ while being vulnerable and after EW had established the spade fit, so the jump was hardly going to be much of nuisance to EW. Therefore the jump had to show a very good hand. Furthermore EW had already 'told' North, that South had a singleton or void in spades.
If East's system had allowed him to bid 3♠ instead of 2♠, the slam would have been practically unbiddable for NS, since 4♥ (not a jump bid now) by South now 'sounds' weaker than it sounded with the jump.

The third and last episode of 'Penalty double or pass' deals with Fredin doubles. Read it here.