Let's imagine a scenario in which a cricket fan is explaining what happened in the fourth Test between India and South Africa recently at the Kotla to someone from outside our charmed circle.

Cricket Fan (CF): It was Test cricket at its attritional best! The Saffers tried to play out ten hours of game time to draw the fourth Test, and their discipline was amazing.

Outsider Dude (OD): What do you mean, "draw the Test"?

CF: It's when neither side wins or loses - we call it a draw.

OD: Oh, you mean like a tie?

CF: No, no, a tie is different. It's very rare. This is just a draw - a no-result. Happens quite a lot in Test cricket.

OD: What would be the point of that? Especially after five days of play?

CF: Umm, never mind. The thing is, South Africa blocked everything the Indians threw at them. They scored at less than a run every over for over eight hours. Their grit was phenomenal.

OD: Hmm, I still don't get it. Were they trying to "draw" the match to come back strongly in the next Test?

CF: No, there is no next Test. The series was already over. India were 2-0 up before the Kotla Test even began.

OD: Wait a sec. Why play a fourth Test if by the third the series was over? You don't have a game five in the NBA finals if one team is up 4-0. Nor does Djokovic play a fourth and fifth set against Federer if he has won the first three.

CF: I don't know why we do that in cricket. The whole series has to be played even if sometimes the last or even the last two Tests are dead rubbers.

OD: So let me get this straight: South Africa played the most incredibly defensive cricket for nearly 10 hours to draw the final match in a series that they had already lost?

CF: Yes. Except, they didn't succeed. They lost the fourth Test too, and so the series ended 3-0 in India's favor.

OD: You know what? This always happens whenever we talk about cricket: I get a headache and am left even more confused about the game than before. Frankly, you guys should think about changing the name of your game to Dead Rubbers - that's the only thing you've said so far that makes any sense at all.

The last couple of weeks have thrown into relief two possible directions Test cricket can go in the future: one spells certain, if gradual, death, and the other a possible resurgence. The first was epitomised by the spectacle at the Kotla, whose sheer absurdity the imaginary conversation above tries to capture. The second is what New Zealand under Brendon McCullum showed in the first Test against Sri Lanka - blazing away at over four runs an over through the match despite losing wickets, and risking defeat with a bold declaration to ensure a result. I firmly believe it's this direction we should be moving in, and it's time to leave the business of drawing meaningless Tests behind us.

In that vein, I'd like to propose a couple of radical departures that are long overdue and have the potential to revolutionise the way Tests are played:

a) Abolish the draw, b) Eliminate the toss, and c) have a limit for how many overs each innings can last. Let's briefly discuss the two changes in turn, realising that a few more tweaks may be necessary to achieve optimal results.

First, from here on, every Test has to end in a result. The team batting last will have to chase down the target to win: if it fails to do so in the available overs, it will have lost the Test, no matter how many wickets it still has in hand. The option of playing out time to achieve a draw no longer exists - you either win or you lose. (Ties are allowed - there will be no equivalent of a Super Over or sudden death overtime to force a result. Ties are likely to be so rare that we can allow their special status to remain.)

Second, toss the toss. Eliminating the draw possibly gives too much advantage to the team winning the toss. To counter that, one of two options may be considered. Teams alternate on who gets to decide whether they are going to bat or field first. In every series, the visiting team gets to decide in the first Test, the home team gets its turn in the second Test, it's back to the visiting team for the third test, and so on. Or, better still, we recognise that most Test series are won by home teams anyway as they are more familiar with the conditions (and often stack them in their own favour) and redress that by allowing visiting teams to always get first pick on whether to bat or field, as has been proposed in English county cricket.

Third, each team can bat a maximum of 100 overs in each of its innings. All Tests will have four days of play beginning at 3pm and ending at 10pm with two 30-minute breaks in between for "tea" and "dinner". In the six hours of play, 100 overs will have to be bowled, at an average of just over 16 overs an hour. Ten more overs a day is not a big addition to the current 90-overs-a-day rule. Bowlers used to routinely bowl even more back in the day. (It's also clear that bowling teams can bowl at a rapid rate when it suits their interests: India averaged 18 overs every hour during South Africa's fourth innings at the Kotla.) Endless fiddling with field settings and a general lack of urgency are to blame for the abysmal over rates we see these days. With a minimum of 400 overs in four days of day-night cricket, the game will move along at a brisk pace.

This combination of an over-limit per innings alongside the abolition of the draw and modifying the toss has the potential to revolutionise Test cricket while retaining many elements that make it unique and special. Teams will still have to put a premium on batting sensibly, as the conditions and bowling warrant, and yet scoring runs briskly will remain important.

Taking wickets as often (and as cheaply) as possible will remain the best way to win a match. You cannot sit on a series lead by preparing batting beauties to ensure draws, and this could also relieve the tedium of dead rubbers. Home-field advantage will be neutered somewhat by alternating who gets to decide to bat or field first, or by giving that option exclusively to the visitors.

Since matches will be played mostly through the evening, live and TV audiences will likely be greater than they are currently. At the minimum, these are ideas worth considering. What we don't need more of is what happened at the Kotla: that's a relic from a bygone era now best left in the past.