Hare & Tortoise is a game of strategy based on numerical calculation and player interaction. But it has always included one built-in element of randomisation in the form of hare squares. The original purpose was to satisfy players who regard pictorial board game as incomplete if they don't include a randomising element, and for this purpose it seemed appropriate to choose one particularly in keeping with the hare's traditional madcap character. The actual mechanics of 'Jugging the Hare' vary from edition to edition for reasons explained below, though most of them have in common the fact that they tend to favour players lagging behind more than those haring ahead. In 2018 Ravensburger asked for an alternative, non-random outcome for use in tournaments, where expert players prefer to exercise maximum control over their carrot-management. Already in some circles competitors omit them entirely, by simply designating them 'unplayable squares' and allowing no one to land on them; in others, they are merely the equivalent of the 'Free Parking' corner of a well-known property trading game. After much experimentation, a new tournament rule has been provided for this feature.

Various methods of "Jugging the Hare" in use hitherto are as follows:

Intellect 1974, Waddingtons 1980

Intellect Waddingtons
In these earliest versions you randomly draw one of six numbered cards and follow the instruction appropriate to the number it shows and your current position in the race, as per the following chart:
1Miss a turnMiss a turnMiss a turnMiss a turnMiss a turn Miss a turn
2Move back to previous vacant carrot squareMove back to previous vacant carrot square Move back to previous vacant carrot squareMove forward to next vacant carrot squareMove forward to next vacant carrot squareMove forward to next vacant carrot square
3Drop back one positionDrop back one position Move up one positionMove up one positionMove up one positionMove up one position
4Chew a carrotChew a lettuceChew a lettuceChew a lettuceChew a lettuceChew a lettuce
5Your last turn free of chargeYour last turn free of chargeYour last turn free of chargeYour last turn free of chargeYour last turn free of chargeYour last turn free of charge
6Play againPlay againPlay againPlay againPlay againPlay again

Ravensburger 1979

Ravensburger 79
Ravensburger preferred to replace the table with a set of cards with universally applicable instructions. You therefore merely drew a card, did what it said, and returned the card to the bottom of the pack. The 12 cards were:
Fall back one position (x2)
Your last turn costs nothing (x2)
Either draw or discard 10 carrots (x2)
Leap ahead by one position (1)
Leap ahead to the next carrot square (1)
Fall back to the previous carrot square (1)
Have another turn (1)
Miss a turn (1)
Chew a lettuce (1)
With no provision for shuffling, players with good memories eventually knew what came next.

Gibsons 1986, Abacus/Rio Grande 2000

Gibsons Abacus
Back to first principles, with eventualities related to your position in the race, but instead of drawing a card with a number from one to six you roll a die, add the number you get to your current position, and follow the instruction associated with their sum:
 2  Move back to the last vacant hare square (if any).
 3  Miss a turn.
 4  Move back to the last vacant carrot square (if any).
 5  Chew a carrot. (Draw 10 out or pay 10 in).
 6  Restore your carrot holding to exactly 65. (If you have more, pay in; otherwise draw out.)
 7  Free turn. (Reclaim carrots paid to reach this Hare square.)
 8  Lose exactly half your carrots. (If there's an odd one, keep it.)
 9  Have another turn.
10  Move forward (free) to the next vacant carrot square (if any).
11  Chew a lettuce. (If you have any lettuces, treat this hare square exactly as if it were a lettuce square.)
12  Restore your carrot holding to exactly 65.

Ravensburger 2008, Gibsons 2010

Ravensburger 2008 Gibson 2010
For Ravensburger's return to the fold I managed to combine the fixed instruction method with making some of them vary according to your position in the race. I also abolished manifestly unfair instructions such as moving backwards or forwards one position, which could otherwise ruin other players' calculations. There were now 15 cards, as follows:
Give 10 carrots to each player behind you in the race (x2)
If there are more players in front of you thane behind, play again; if not, miss a turn (x2)
Your last turn was free (x2)
Lose half your carrots (x2)
Restore your carrot holding to exactly 65 (x2)
Draw 10 carrots for each lettuce you still hold. If none, miss a turn (x2)
Reveal your carrot holding to everyone else (x2)
Shuffle the hare cards and receive 1 carrot from each player for doing so (1)
The same system is followed in Gibson 2010.

Devir Iberia, 2014

Devir Iberia (2014) found that the cards wouldn't be big enough to contain written instructions in three different languages (Spanish, Portuguese, and Catalan). Instead you draw the top card of a shuffled pile, which tells you (in numerals) to draw or pay in 5, 10, 10, 15, 20, 20 or 25 carrots, or, in one instance, to shuffle the cards with neither gain nor loss, thus preventing players from remembering a sequence. Unfortunately, this loses the feature of varying the outcome according to your position in the race.

The new tournament rule

When you land on a hare square do nothing immediately. If nobody goes past you, draw 10 carrots when you next move. If they do, pay 10 carrots into the carrot patch for each player who goes past you, in either direction.
The strategy of Hare & Tortoise derives largely from player interaction, in that players can change their opponents’ access to carrots by overtaking them, backwards or forwards, when they are on a numeral square. This new tournament not only conforms to this principle but also continues the tradition of favouring a trailing player over a leading one. In effect, the hare square becomes a carrot square, except that its consequence now depends upon others overtaking you or not instead of giving you a free choice. It's also a very simple rule and easy to explain.