I was a little sore at Prof. Bainbridge for using my post on Alito as a punching bag in a way that would be off the mark if he had read the whole thing, so I held off responding until now. I'm also impatient with the straw man dichotomy between judges who respect the law meticulously and judges who careen in all directions recklessly doing whatever they want, as though those distinct types comprise the universe of choice. Mark's post below demonstrates one central way in which even a very conservative judge's fundamental beliefs will affect his decisions. But there's more to this.
In the first place, cases that come to the Supreme Court are hard cases, meaning that legitimate values (including values set out abstractly in laws like the constitution) and settled laws conflict. Decisions may turn on whether the value of stable expectations ("an old tax is a good tax") or protection of a constitutional right ("a poll tax for voting is not a good tax") should count more to decide the case at hand, and no parsing the technical record of law will answer this, any more than scrutiny of the nutrition label will help you decide whether to make your kid's birthday cake from chocolate or vanilla mix. The kind of values justices hold in especially high regard will tilt decisions on the whole one way or another, and it's entirely appropriate for the president and the senate to try to read these values and use them in making appointment decisions.
Prof. Bainbridge does corporate law, a context in which the parties are usually not people, have no human rights, and cannot suffer (I know, shareholders and employees figure in it) and in which the issues for all I know can always be settled by reference to statutes. But among the laws important to appellate practice is the constitution, whose preamble says its purpose is to "establish justice", an abstraction that simply cannot be used without interpretation and specification for cases. At one time, that word meant a defendant should only have to put his hand in the boiling water for a few seconds to see if he was guilty - that it would be unjust to cook him until his flesh fell off the bones. Now we have a different idea, Miranda rights and such like. But what it means in practice will always depend on its expression in the mental constitution of judges, an expression that cannot be blank, nothing, or labile and contingent on the case at hand.
The idea that a decision must properly flow only from the enacted statutes (the unique expression of political will) of a state and not from the contradictory holdings of lower courts which are to be resolved, or precedents of higher courts, is the foundation of the Roman Law model that contrasts with the Common Law system used in Anglo-Saxon countries, including ours. It may or may not be a better system, but we haven't bought it; in any case, it doesn't work the way it's supposed to even where it lives, as the immortal 1976 Hastings Law Review article by Prof. Merryman ("The Refrigerator of Bernard Buffet") demonstrates.
It is also impossible to decide the big cases properly looking through a knothole only at the law. Even tough conservatives seem to think Brown was decided properly, and that holding depended greatly on sociological research by academics. In using evidence like this, the relative weight of different moral values of the justices has to matter (I don't mean this normatively, I mean it positively): my problem with Alito is simply that the moral value of human rights and justness seem to be fairly far down his salience scale compared to other things that matter to him. This will matter when he votes on close cases, and it's entirely an appropriate concern of the Senate and the public.