The article states "[The standard model is] the most accurate theory ever developed, in any field." I would be fascinated to know the objective standard to which models in different fields can be evaluated for "accuracy". Certainly, all models are wrong, but their value comes in helping make predictions which turn out to be correct. Is the accuracy of a theory dependent on how many correct predictions it has made? As measured by what...number of publications? And when is a theory so uncontroversial that it ceases separating one group of thinkers from another? Natural selection exists largely (as a theory) because it stands in opposition to other schemes of explanation.

Now, if you're working "accuracy" from the "nothing left underlying the current explanation" angle, then the science of physics is always going to win because of a very uninteresting fact. Science is integrated. We carve it up into areas to help ourselves, not because Nature is divided up into sciences. (Psychology is constrained by biology is constrained by genetics is constrained by chemistry is constrained by physics.) No theory in any other area is complete without the proper foundational work in all of the science which supports it (and physics supports everything). If that's what "accurate" means, then shove off...the claim is meaningless. The best theory in physics will always be the best theory ever (even if it's crappy) because bad physics necessarily takes the rest of (fully articulated) scientific theory down with it.

Mangar:

I would be fascinated to know the objective standard to which models in different fields can be evaluated for "accuracy". Certainly, all models are wrong, but their value comes in helping make predictions which turn out to be correct. Is the accuracy of a theory dependent on how many correct predictions it has made?

This claim refers to the precision (as measured by number of decimal places) to which predictions of the theory have been experimentally verified, as compared to any other theory in any field that produces quantitative predictions. It does not mean the theory is more accurate than, say, a theory that predicts that the sun will rise tomorrow, which would be, in some sense, "infinitely accurate."

You seem to regard some very weighty and complex philosophical questions (whether models are or can be true, whether science is unified, etc.) as obviously settled. You might be interested to investigate some of the literature on these topics.

And indeed, the results of experiments in particle physics agree to theoretical predictions to an amazing degree of precision.

The theory may, however, be completely inaccurate.

Well, I'd be happy to read anything you'd like to write on the weighty and complex topics at hand. Please don't assume that my take on these topics is uninformed because I don't dwell on them.

I appreciate the "precision" argument as a bit more constrained than the "accuracy" argument, but I still have a hard time feeling like it does the necessary work in comparing theories across fields. Seems like "number of decimal places" depends on your units, and the value of a theory depends on the width of its application (a theory constrained to the effects of one or two causes could be quite precise, as long as the discounting of other causes was written into the scope of the theory.) "Number of decimal places" might be a fine, fine way to evaluate theories in physics, but my point is that benchmarks for accuracy, precision, usefulness, etc. will be different in different areas of science because we view (and measure) different areas differently.

It might be easier to compare social science to physics if we only considered models of variance explanation...that is, the differences between people on, say, intelligence, (about which I will make up random number) is 78% explained. That leaves 22% of the variance to error in the model, and it can be compared to other models which can calculate an error term (the error term can never be completely eliminated, which is the underlying philosophy behind my "All models are wrong" statement). However, not everything interesting is about variance, and a model which seeks causes for universal traits (where there is no variance) will not be amenable to such anayses.

Schadenfreude:

The theory may, however, be completely inaccurate.

If you think it makes sense that a theory whose predictions are confirmed to an arbitrarily high degree of precision might nevertheless be false, you are probably tempted by some flavour of the philosophical position of anti-realism. I'm not unsympathetic to that, but that argument is not going to be part of most scientific discussions of the merits or accuracy of theories.

Mangar:

I appreciate the "precision" argument as a bit more constrained than the "accuracy" argument, but I still have a hard time feeling like it does the necessary work in comparing theories across fields.

I have no objection to that. I'm not committed to making sure the Standard Model of particle physics (or any theory) retains the title of "most accurate," or even "most precise," for that matter. I doubt it is useful or informative to compare theories in different fields this way beyond a pretty superficial level. Theoretical physicists aren't particularly hung up on it, either, I don't think. This claim gets repeated mostly just to underscore the predictive power of the theoretical apparatus developed and the astounding experimental feats that have tested it.

> Seems like "number of decimal places" depends

> on your units

Not if it's a ratio (the number of digits would depend on the base, like base 10 or binary--but that's trivially interconvertible).

I wish I could enlighten you on what the ratio is, but I don't know. Perhaps someone can link us to a Wikipedia etc. article? I'm sure this is discussed. In any case, I believe the agreement between prediction and measurement is on the order of eight or ten decimal digits.

By "most accurate", I think they mean precisely that. The standard model predicts values that agree with experiment up to 8+ degrees of precision. Experimentally, we have determined the values of the free parameters of the SM to high precision. Now, using these, we then calculate what we might expect from an experiment theoretically, and the agreement is astounding.

When you are dealing with millions and even billions of events, it becomes easier to obtain this accuracy. If I were studying vortices created by helicopter blades, I would have to create probably 100 to 1000 to fully understand the interactions involved. When we look at particle physics experiments that study the SM (not including neutrino experiments), we have many orders of magnitude more events to study. These sheer numbers of events allow us to confirm that the SM is, in fact, the most accurate theory thus created. (I refer any interested person to pdg.lbl.gov, which is the Particle Data Group website, listing everything anyone could want to know about the SM and fundamental particles in general.)

As for the article itself - I think it really comes from a lack of understanding of the current stance of physics. Looking up the Standard Model on Wikipedia (something the author could have done easily) would have resulted in a clearer idea of all the physics that will be examined at the LHC. If we include the Higgs, we still can't even explain Dark Matter, nor the massive Baryon Asymmetry, nor any number of other issues there are with the SM. Particle Physics don't suddenly go "ah hah! we found a five sigma deviation! Well... that was anti-climactic". There is far more to studying the Higgs than just discovering it. If the Tevatron has found it, I am happy. But it doesn't prevent me from still looking forward to the results from the LHC as well.

Particle Physics is not a competition between involved parties in the same way as capitalist corporations compete. Sure, everyone wants to make a name for themselves and make the big discoveries. But on a more fundamental level, it is about a cooperation to advance physics and understand the nature of matter and the Universe. The ATLAS group, for example, might be a little upset that they didn't get to discover the Higgs first (if the Tevatron rumor turns out to be true), but I assure any reader that those involved aren't sitting on their thumbs with nothing to do any longer. They are going to be rushing to change their plans.

Experimental confirmation will be an important next step, as will further examination of the properties of the Higgs. Its a long process, and the LHC will not be wasted in any way by any discovery from the Tevatron.