Some days you get to be a grand visionary. This being October, you’re more likely to be busy preparing your budget.

Budgeting is a problem. It’s a time-consuming, annoying, expensive process, almost always implemented as stupid game-playing (the game is Pin the Tail on the Donkey), and the only thing worse is not doing it.

One difference between budgeting and other games, like bridge or Scrabble, is that you can choose to not play bridge and Scrabble. If you want to be in management, you have to learn how to play the budget game to win, or failing that to play it to not lose. It isn’t easy, since the other side holds most of the good cards.

(Sidebar: A group called the “Beyond Budgeting Roundtable” is trying hard to replace the budgeting game with a more rational approach, built on two major pillars — continuous, adaptive planning and decentralized decision-making. I wish them well. If your company hasn’t adopted Beyond Budgeting, though, I don’t recommend that you be the first to say, to the CEO and CFO, “You know that our budgeting process is a stupid game that promotes unethical behavior. Why don’t we adopt this promising alternative?” Unless, that is, you’re seriously interested in committing career suicide.)

There is no single, magic formula for preparing and successfully defending next year’s IT budget. Here are some techniques that should help you survive the experience, and even to have intelligent conversations on the subject:

• Separate spending for value maintenance from spending that enables new business value. Value maintenance is “non-discretionary” — curtailing it isn’t a realistic choice. So long as the company is in business you will run day-to-day operations, modify the accounting software to accommodate changes in the tax code, and implement new labor contracts in the payroll system.Value enablement, in contrast, is “discretionary”: The business has a choice about each proposed software enhancement and major project, and about business expansion that drives the need to invest in additional infrastructure — for example, opening new branch offices.

• Base spending for value maintenance on ratios. The cost of maintaining already-delivered value is the result of some basic drivers, such as the number of servers each sysadmin can support and the cost per seat for software licenses. Build your budget on these ratios, so you can defend the need to increase non-discretionary spending if the business situation calls for it.If you can, add another layer of sophistication by going beyond ratios, to break down costs into fixed and variable components.

• Build continuous improvement into spending ratios. When you can, of course: You can’t build continuous improvement into per-seat software license costs. You can, in contrast, take steps to increase the server/sysadmin ratio, and you should, if for no other reason than to prevent complacency from seeping into your organization. Improving these ratios is how you shift spending from value maintenance to value enablement without reducing the value you maintain in the process.

• Decide whether value-enabling spending is driven by capacity or demand. If you’re capacity driven, it means governance decisions are purely about allocating the staff you have available to work on software enhancements, major projects, and business expansion. The budget decision is about how much staff you have available to allocate.If your spending is driven by demand, it means the company decides what it will take on and you provide the right mix of employees and outside resources to take care of it. That, in turn, means you can’t budget without the company’s plan. It also means you have to be very good at the discipline of project cost estimation.

• Apply the “walkaway rule” to your budget negotiations. For better or for worse, in most companies budgeting is a negotiation, not a collaboration (okay, I lied; it’s almost always for worse). One of the most basic rules of negotiation is that whoever is in a better position to walk away from the deal wins.In practical terms, this means agreeing to whatever cuts in your budget are proposed by the other side, making two points clear in the process: (1) the business will have to do without something as a result of the cut; and (2) you won’t be the one in the headlights — the standard governance process will be the mechanism through which the company decides exactly what the business will have to do without. That’s what it’s for.

The only thing worse than having to play stupid games is losing stupid games you’re forced to play. You might as well play to win.