Louisiana elects all its state legislators in the odd year just before presidential election years. All state legislators in both houses get four-year terms. The last election for Louisiana legislature was in 2007.
Louisiana has used “top-two” in all state elections starting in 1975. In the 2007 election, out of all 144 legislative races, only two incumbents were defeated for re-election in either round of voting. They were Rick Farrar, a Democrat in the 27th House district, and Carla Dartez, a Democrat in the 51st district. Both were defeated by a Republican.
Louisiana has term limits. So even though no incumbent senators were defeated, 18 of the 39 senators were new.
So instead of 39 senate races, 39 incumbents are re-elected (as one might infer from the original post), it was actually:
21 senators seeking re-election:
8 unopposed;
12 re-elected in the primary;
1 re-elected in a runoff.
In 10 of the 13 seats in which an incumbent was challenged, there was no opponent from the opposite major party, which suggests part of the reason that they were re-elected was the nature of their district.
Most were re-elected quite handily, 9 with over 2/3 support.
The incumbent re-elected in a runoff defeated a former incumbent that the current incumbent had upset 4 years previously in a runoff.
Term limits may also have an effect of limiting strong challenges to incumbents. The best positioned challenger may be similar to the incumbent – presumably there is some correlation between the demographics and political leanings of a district, and the person that is elected to represent the district. Rather than attempting to defeat an incumbent who may have similar political views, ethnicity, etc., it may be wiser to become their legislative assistant, and run as a quasi-incumbent when they are term-limited in 4 years, with their endorsement and black book of campaign backers.
There were 18 open seats.
1 newcomer was unopposed;
8 were elected in the primary;
9 were elected in a runoff.
5 of the races decided in the primary only had two candidates, but were decided by mostly modest majorities: 69, 63, 54, 52, and 51%. The other races had 3 candidates each, and the victor had 64, 61, and 51%. Two of these races had 3 candidates of the same party. Under the corrupt partisan primary system now used for congressional elections in Louisiana, these would have been decided by only party members, in Louisiana this effected by having election judges manipulate lockout devices on the side of voting machines before permitting a voter to vote. While they were certain to elect a candidate from the dominant party, all voters were allowed to vote.
5 of the 9 open seats that went to a runoff were dominated by one party or another, and both candidates in the runoff were from the same party.
In the two of these 5 that where there was a candidate of the minority party, that candidate and party had limited support. The Republican in S14 finished 3rd with 8% of the vote; and the Democrat in S37 finished 4th with 17% of the vote. Again all voters in the district were permitted to vote in both the primary and in the runoff. Under a partisan primary system, adherents of the minority party would have been blocked from voting in the 1st primary and 2nd primary, and only allowed to support their party’s candidate in the general election where they might have been defeated by a 20-80 margin.
In the 3 districts where all 3 (or 4) candidates were from one party, minority party supporters would have been barred from the 1st primary, the 2nd primary, and, in Louisiana, the general election would have been cancelled.
In the other 4 races, where there was a Republican vs Democrat in the runoff, there was a 3rd candidate from one or the other of the two parties. There appears to have been some cross-over voting since the candidate who was the sole candidate for his party gained 6/19, 15/32, 8/30, and 6/9 of the eliminated candidate’s support (eg in S17, the candidates finished DRD in the primary, with the eliminated Democrat receiving 19% support. In the runoff the Republican candidate increased his support by 6%).
There were at least 4 senate incumbents defeated in 1995, 2 in 1999, and 1 in 2003.
Defeat of incumbents may not be a good indicator of the quality of an electoral system. Consider if an election is the equivalent to the hiring of a new employee (ie the voters are hiring their representative or even an executive). If a few years later, they replace the employee, what does it indicate? Perhaps the original selection process was not so good. But rather than correcting a mistake, by replacing an incumbent, the system is careening from mistake, haphazardly making selections. There is of course a possibility that the working conditions have changed – new issues have arisen, or the electorate has changed through redistricting or population shifts. But with 3-term limits on service in Louisiana, there is relatively short amount of time for this to happen, and redistricting occurred before the 2003 election, so that 2007 was the 2nd election on the current boundaries.
Only 21 of 39 incumbent senators and 46 of 107 incumbent representatives sought re-election.
So rather than incumbents being beaten in 2 of 144 races as some might infer; incumbents were defeated in 2 of 67 races where they sought re-election. I don’t see what the 1975 initiation of the open primary has to do with the number of incumbents that were defeated. You didn’t present any data that indicated that the number of defeated incumbents was less, more, or the same over the past 32 years. Looking back at the last couple of elections for which data is available on the SOS web site suggests that 2007 was somewhat low in terms of incumbent upsets, but that may be entirely due to the relatively small numbers of incumbents seeking re-election.
In the House the turnover was quite remarkable. There were 61 newcomers at the start of the session in 2008, versus 17 in 2004. 11 of the 44 holdovers in 2008 had themselves first been elected in special elections after the 2003 general election, so only 33 of 105, or 31% had completed 1 full term.
Because of the massive turnover, the percentage of incumbents defeated in a re-election attempt was quite similar between 2003 and 2007 (5 of 93 or 5.3% in 2003; vs. 2 of 46 or 4.3% in 2007).
Some more details:
46 incumbents sought re-election:
19 were unopposed;
24 won in the primary, while 1 lost.
1 won in a runoff, while 1 lost.
So not only were 44 of 46 incumbents who sought to return, returned, most were easily returned. Of the 27 incumbents who actually faced a primary, 14 were in 2-way races. So rather than a serious challenge, they may have only faced a candidate who filed (In Louisiana, if only one candidate files for an office, the election is cancelled). 10 incumbents faced and defeated two opponents. 3 incumbents faced were in 4-candidate races. Two of these required a runoff, including the runoff that was lost.
One incumbent avoiding being defeated by 33 votes out of 14,697 votes by an independent challenger. If she had won, it would have been the 3rd independent elected to the House. One independent was an unopposed incumbent elected in a special election during the previous term, and the other won in an open seat.
There were 59 open seats.
7 of the races had only a single candidate.
14 races were determined in the primary.
38 races had no majority in the primary.
Of these 38 races, 34 were decided in a runoff. In the other 4 races, the 2nd place withdrew. All were in extremely Republican districts, where the Republican candidates collectively had 100%, 89%, 89%, and 82% of the votes; the primary leader had near a majority with 49%, 44%, 44%, and 42%; and had a 20% plurality over the 2nd place candidate, another Republican.
There were an average of 3.46 candidates in the open seat races, and 3.79 in the races that were contested.
7 had one candidate, and the primary was cancelled.
11 had 2 candidates, and were won in the primary.
15 had 3 candidates, with 13 going to a runoff.
12 had 4 candidates, with 11 going to a runoff.
10 had 5 candidates, with 9 going to a runoff.
2 had 6 candidates; with both going to a runoff.
2 had 7 candidates; with both going to a runoff.
1 had 8 candidates; and went to a runoff.
13 of 15 open seat 3-candidate races, and 38 of 42 with 3 or more candidates required a runoff. Compare this to races involving an incumbent. 0 of 10 those 3-candidate races, and 2 of 13 3 or more candidate races went to a runoff.
As was seen with the senate, relatively few incumbents are challenged. But open seats are truly open, with all voters permitted to participate in the selection of their representatives.