Over the past two years, dozens of politicians and prognosticators have drawn various red lines that Iran should not cross lest it be “a screwdriver turn away from having a nuclear weapon,” as Senator Bob Corker, the Tennessee Republican, said last week.

Mostly they focus on centrifuges, the water-heater-sized machines used to enrich uranium. You can understand why. Centrifuges are part of the elaborate process used to turn uranium ore into the metal core of atomic bombs. They are perhaps the most quantifiable part of the process. They are discrete objects that can be numbered.

And that is what we do. We count things. It is one of the first skills we teach our children. It helps us put a little order in the universe. How many kids in the classroom? How many votes to elect a president? How many stars in the sky?

We can easily count centrifuges. Anyone with a computer can come up with an estimate of how many centrifuges Iran needs to make the material for a bomb. Just search Google for “Iranian centrifuges.” The very first hit is an article produced by the publication Iran Watch that “estimates how soon Iran could fuel a nuclear weapon.”

By using the approximately 9,000 first generation centrifuges operating at its Natanz Fuel Enrichment Plant, Iran could theoretically produce enough weapon-grade uranium to fuel a single nuclear warhead in about 1.7 months.

There you go. To limit Iran’s weapon potential, cut the number of centrifuges. No more than 10,000, say some. No more than 4,000, say others. Between 3,000 and 4,000, say still others. Even 1,000 could be too many, claims Harvard's Olli Heinonen.

It is a simple metric for success. And it is wrong.

There are, in fact, many ways to limit Iran’s ability to make a nuclear weapon. Centrifuges are just one factor in the equation.

* * *

There are multiple, industrial steps in the enrichment process, including mining the uranium ore, purifying it into a powder known as yellow cake, mixing that powder into a gas and then spinning that gas in centrifuges.

The centrifuges increase the ratio of the fissile isotope, Uranium-235, from the natural ratio of less than 1 atom in 100 to about 5 in 100, or 5 percent enriched uranium. At about that concentration, the U-235 atoms are close enough together that they can sustain a chain reaction.

You can stop the process there, turn that gas into powder again, process the powder into fuel pellets, form the pellets into rods, insert the rods into a reactor, and use the heat from the fission to turn water into steam that spins turbines, generating electricity. About 20 percent of the electricity in the United States is produced in exactly that way.

The problem is that the same centrifuges that enrich uranium for fuel can also enrich the uranium for the core of a bomb. With some reconfiguration, the same centrifuges can keep going to 70 or 90 percent U-235, or weapons-grade. At that concentration, it only takes about 50 pounds of the material to ensure that a single neutron hitting a single atom will trigger an uncontrollable chain reaction, unleashing in a microsecond enough energy to destroy a city. That is why centrifuges are so important.

Most countries that have nuclear power reactors do not have centrifuges. They buy their fuel from the handful of countries that make it, including the U.S., Canada, Russia, and a European consortium known as URENCO. Russia, which is constructing Iran’s power reactors, is happy to sell Iran the fuel and dispose of it when it is spent. But Iran says it wants to make its own fuel to ensure a steady supply. The question is: Do you trust them? Clearly, we do not.

The deal now being negotiated between Iran and the six countries known as the P5+1 (the U.S., Great Britain, France, Russia, China, and Germany) will reportedly cap the number of centrifuges Iran is allowed to keep. Israeli officials appear to have leaked to the press confidential information provided them by the U.S. that places the number at around 6,500 to 7,000 centrifuges. This would be a sharp drop from the 20,000 machines Iran now has. But that is still too many for Israeli Prime Minister Benjamin Netanyahu, who is demanding zero centrifuges.

Zero is a fantasy, and you can blame President George W. Bush’s administration for that. It may have been possible to convince Iran to dismantle all its centrifuges when it had only a few dozen in 2003 and first offered to talk to the U.S. Or in 2005, when it had a few hundred and was in talks with the European Union. But the Bush administration spurned any deal. “We don’t negotiate with evil,” said Vice President Dick Cheney. “We defeat it.” As a result, the talks collapsed and Iran went from zero centrifuges installed at the beginning of the Bush administration to about 6,000 at the end.

There is not a political leader in Iran today that could agree to completely dismantle its nuclear fuel complex. But some, including Iranian President Hassan Rouhani, seem ready to sharply limit it. The key to a solid deal is to couple limits on the number of centrifuges with other limits that prevent Iran from quickly building a bomb should it break the deal.

* * *

The easiest way to do this is to limit the quality and amount of uranium gas that Iran has to feed into the centrifuges. Netanyahu, with his famous cartoon bomb at the United Nations in 2012, warned that Iran was near his red line because it would soon have enough 20 percent-enriched uranium gas to feed back into the centrifuges and produce enough highly enriched uranium, or HEU, for a bomb. Some experts warned in early 2013, “We estimate that Iran, on its current trajectory, will by mid-2014 be able to dash to fissile material in one to two weeks unless its production of 20 percent-enriched uranium is curtailed.”

Its production was curtailed. This is no longer a threat.  The interim deal negotiated by the P5+1 in November 2013 effectively drained Netanyahu’s bomb. Iran has eliminated its stockpile of 20-percent uranium gas and has stopped making any more.

But it still has over 8,000 kilograms of uranium gas enriched up to 5 percent purity. If Iran were to feed that gas back into its operating centrifuges, it would theoretically take between 2 to 4 months to refine it into enough HEU gas needed to make the core for one weapon.

A solid deal would greatly reduce the amount of uranium gas Iran is allowed to keep on hand. It would also prevent Iran from replacing its current, inefficient model of centrifuges with newer designs, limit the production capabilities of the existing cascades and put in place tough, new inspection regimes that could detect any cheating. Experts at the Arms Control Association estimate that:

By reducing Iran’s current operating enrichment capacity by half, combined with a significant reduction in the size of Iran’s low enriched-uranium stocks and conversion to more proliferation-resistant oxide form (or removal to a third country), the time it would take Iran to produce enough weapons-grade enriched-uranium gas for one nuclear weapon would grow to nine to 12 months.

The goal of such limits, as former State Department official and Iran negotiator Robert Einhorn explains, is to ensure that:

… the timeframe between the initiation of breakout and the production of enough weapons-grade uranium for one or more weapons is as long as possible; and that once breakout is detected, the international community will have the will, the capability, and the time to take effective action, including the use of military force, to prevent the acquisition of enough fissile material for a nuclear weapon.

So, how much time?  This is something else we can count. In general, critics of the negotiations have insisted that there be at least 6 to 12 months time.

Do we really need a year to respond?

Mark Fitzpatrick, nonproliferation director at the International Institute for Strategic Studies says no:

The breakout factor is not the all-consuming issue in London that it seems to be in Washington. In the negotiations, the British go along with the other members of the [P5+1] on the timeline calculations, but are realistic about the human factors and operational troubles that Iran would experience if it tried to produce 90 percent HEU. Given these practicalities, what might be characterized on paper as a six-month breakout period in a negotiated deal would actually be longer.

In other words, estimates of breakout times measured in months or weeks assume that everything goes right and nothing goes wrong—a condition that rarely exists in the real world. Many, many things have gone wrong in the Iranian program over the decades; nuclear research began over 60 years ago during the time of the Shah.

Even in the worst case, Iran would just have the material for one bomb. It could take an additional 6 months to a year to turn that material into an actual weapon. But it would still have just one weapon. No country has ever “broken out” with only one weapon, raising serious doubts about this entire scenario.

Former CIA analyst Paul Pillar criticizes “the fixation” on breakout times. “The difference between, say, six months and a year is meaningless,” he says, “when any conceivable response, including military attack as well as enactment of the most debilitating possible sanctions, could be accomplished within a couple of weeks.”

* * *

Could we actually detect and act on a breakout in that short a time? Currently, inspectors from the International Atomic Energy Agency, or IAEA, are allowed daily access to Iran’s enrichment facilities, thanks to the interim agreement. Inspection procedures will certainly intensify under any final deal.

With inspectors, cameras, seals, inventory controls, tracking of scientists, monitoring of production from the time uranium is taken as an ore from the mines to when it is stored as a gas in cylinders, any move to switch from producing low- to high-enriched uranium would likely be detected within a day or two. It would take another day or two to report this to the IAEA, a few days for the board to convene and pass a resolution, then another few days for the UN Security Council to engage. Total time could indeed be a couple of weeks.

But the global response is just one of the factors deterring Iran from violating a final deal. If Iran were caught cheating, “U.S.bombers could be overhead to stop it within 24 hours,” says Fitzpatrick, “This is a strong deterrent against such a breakout option. So there is no practical difference between a deal that provides for a six-month breakout period and one that provides for 12 months.”

It is not breakout that should worry us. It’s the breakdown. It appears that Netanyahu is trying to engineer a collapse of the talks. He is pushing the Senate to adopt new sanctions on Iran that the head of Israeli intelligence said would be “like throwing a grenade into the process.” The idea, according to some supporters of this plan, is to cause a crisis in the negotiations, so that the deal now being discussed would collapse and talks would resume some time later, presumably with Iran in a weaker position.

Some don’t want any deal at all. “The end of these negotiations isn’t an unintended consequence of congressional action,” admitted Senator Tom Cotton, the Arkansas Republican. “It is very much an intended consequence—a feature, not a bug.”

The breakdown of the talks at this stage would lead to the worst of all possible worlds. It would repeat the flawed strategy of the Bush administration. Rather than increasing pressure on Iran, it would decrease it. The U.S. would be seen as the reason for the collapse. Global support for the international sanctions regime would wither. Restraints on Iran’s commerce would dwindle and its oil sales and revenues would climb. The interim deal would be dissolved and Iran, freed of restraints, could increase its production capabilities without limit.

Iran could increase its supply of enriched uranium and bring online more and newer centrifuges. It could move ahead full speed on a plutonium production reactor that provides another pathway to a bomb. And, without inspectors in the country, it might build new secret facilities, opening up yet another pathway. Its theoretical breakout time would go from the current two or three months to two or three weeks.

In short, as Ilan Goldenberg and Robert Kaplan point out, the United States would be forced “to choose between two terrible options both of which are much costlier than the status quo—pursuing military action against Iran or accepting a nuclear-armed Iran.”

It appears that an agreement to verifiably confine Iran’s nuclear program is within reach. It could put Iran’s nuclear program in an iron box with a camera on it. It is not without risk, but every deal has risks.

Whatever uncertainties a negotiated agreement may bring would still be far more manageable than the uncontrollable consequences of a new war in the Middle East and an unconstrained Iran.