Jack Anderson's article "War Toll Estimate: Up to 30,000 GIs in 20 Days" {Nov. 1} was full of inaccuracies and misunderstanding of the realities of modern armed conflict.

Mr. Anderson said, "Top secret Pentagon estimates in the hands of President Bush tell him to expect 3,000 to 30,000 Americans to die in the first 20 days of a war against Iraq."

The figure of 3,000 dead is fairly close to the mark, but a little high; the figure of 30,000 is ridiculously high.

I estimate American casualties in a 10-day battle to liberate Kuwait at about 8,000, including about 1,200 killed. The casualties in a battle depend (among many other considerations) primarily upon the scenario. I do not presume to have enough detailed information about the situation and the forces in the Gulf to suggest that my forecast is more than indicative. I have no doubt, however, that my estimated figures are accurate within 100 percent -- meaning that the maximum casualties U.S. forces are likely to incur in a 10-day battle to liberate Kuwait would be about 16,000, with a maximum of 2,500 dead.

Mr. Anderson stated, "As a rule of thumb, an attacking army should have a 3-to-1 advantage over its opponent in number of troops. But in this case Iraq would be on the defensive with a 4-to-1 advantage over the United States."

There is, however, not much basis for the 3-to-1 rule of thumb. Any substantial collection of reliable battle statistics will show that the average strength of a successful attacker over a defender in wars of this century has been slightly less than 2 to 1. In fact, in 1967 and 1973 the Israelis attacked successfully in most battles with a slight numerical inferiority, less than 1 to 1.

As to the 4-to-1 advantage of the Iraqis over the American forces now in the Gulf, that would presume that Saddam Hussein would move troops from his borders with Iran and Syria, which is most unlikely. Further, it does not account for the substantial number of Arab, British, French and other allies in the combat region and ignores our tremendous superiority in the air and in naval forces along the coast.

I estimate that the relative combat powers of what we and our allies can bring to bear against the Iraqis in the next month would be slightly more than 2 to 1 in our favor. My experience in analysis of historical battles tells me that any combat power ratio of more than 1.5 to 1 is overwhelming. T. N. DUPUY McLean

The writer, a retired Army colonel, is the author of several books on military strategy.