THIS is a question I think we ought to consider this early on.
In a democracy, it’s the will of the majority that decides to vote, that counts.

The other interesting thing is, is it premature to write the obituary on the command vote? In 1998, I wrote that as far as presidential elections were concerned, the days of party machinery determining the outcome were over (the aberration would prove to be 2004: and that was a questionable election).
This election marks 100 years since we’ve had lower house elections; and 66 years since the first national senatorial election. We are only a young democracy in terms of our personal memories.
I’ve put together a summary of the elections from 1907-2004, and it is in the context of all these past races that the present one should be considered. Please take a look (unfortunately, 1971 is a kind of “ghost year,” I’ve never been able to find the House results for that year, though we have the Senate results).

(Erratum: a reader in my blog corrected the data for the 1951 senatorial elections: it was a complete defeat for the incumbent’s party; something pointed out in an Inquirer editorial as the only instance a total opposition victory in the Senate has been achieved; so for the data, the Senate results for 1951 should show 8 NP elected and 0 LP; the above’s been fixed)
In terms of the House of Representatives, the administration in power has not been the same as the party controlling the House only four times in our history, and all three were presidential election years. All three were unusual elections: the incumbents lost, though the party infrastructure the losing incumbents had carefully nurtured survived (until raided by the successor):
In 1953, when Magsaysay won the presidency for the opposition NP, while the administration LP kept control of the House.
In 1961, when Macapagal won the presidency for the opposition LP, while the administration NP kept control of the House.
In 1965, when Marcos won the presidency for the opposition NP, while the administration LP kept control of the House.
In 1998, when Estrada won the presidency for the opposition LAMMP, while the administration Lakas kept control of the House.
All four presidents engaged in the immediate courting and raiding of the previous administration party, and quickly established a “new” majority for themselves.
But for non-presidential election years, no incumbent has ever lost control of the House, and so it’s no achievement to retain control of it in a mid-term election. You could say no administration has ever lost the House in 100 years, that’s simply how local politics works (incumbents have lost the presidency though).
In terms of the Senate, from 1941 to 1951, bloc voting was in place and ensured administration control of the Senate in every election. After 1951, bloc voting was abolished and the results began to be mixed, with one exception: 1955, the Magsaysay mid-terms, when his massive popularity secured a complete senate victory for his slate, also the last time ever that a party secured a complete victory in the senate. Even Marcos at his most formidable, in 1967 and 1969, and even with the anti-Marcos backlash in 1971, neither party could secure a shut-out.
In terms of mid-term elections being a referendum on the sitting administration, the score cards of various presidents is as follows:
Roxas, 1947: 7 out of 8, very impressive
Quirino, 1951: 8 out of 8, very impressive (but led to abolition of bloc voting) All 8 of his candidates lost, the worst performance by an administration, ever.
Magsaysay, 1955: 8 out of 8, hugely impressive and unmatched since
Garcia, 1959: 5 out of 8, unimpressive, a sign of defeat to come
Macapagal, 1963: 4 out of 8, a sign of defeat to come
Marcos, 1967: 7 out of 8, very impressive
Marcos, 1971: 2 out of 8, worst showing of an incumbent; panic time
Ramos, 1995: 10 out of 13, very impressive
What sets apart the pre-1972 senatorial midterms, from the midterms since 1987, is of course that presidents prior to 1972 could run for reelection, and so the midterms also served as a test of how the incumbent would do when seeking reelection. Roxas and Magsaysay were widely expected to secure reelection in their time: but both died before they could do so. Marcos, who matched the Roxas performance in his mid-term, became the first president since 1941 to win reelection.
Since 1987, what the mid-term determines is whether the president will be a lame duck or not. Ramos’s strong senate showing helped propel charter change efforts; 2001 would have been the mid-term referendum on Estrada, but instead became a referendum on Edsa Dos; now the 2007 elections will determine if President Arroyo will be a lame duck or not.