1. Background
Prior to the 13th General Election (GE13) we came up with a methodology of predicting election results based on voting patterns in previous elections (reference).
Our method relied on mapping polling lane results to individual voters. This process assigned probability values (chance of turnout; chance of voting for each coalition) to the voter that was not affected if they migrated to another constituency. This is important because between GE12 and GE13 527,849 voters migrated to different constituencies.
The impact of voter migration cannot be measured for a single seat just by comparing results of GE12 and GE13 for that seat. An analysis of the whole country needs to be performed. New voter registrations, voters passing away and voters no longer eligible to vote are other factors that require deep analysis.
After GE13 we were able to apply the same estimation method to voters based on GE13 results. By comparing the shift in probabilities we are able to calculate the swing in support for each coalition. Because we base our calculations on individual voters, we are able to calculate shifts in support based on combinations of the following dimensions:
- By Age
- By Race
- By Gender
- By Urban Development Category (rural / semi-urban / urban)
- By Parliament/State Assembly Seat
- By Polling District
- By Locality
- By Seats Won by Specific Parties
Any voter whose level of support cannot be determined is assigned a probability of 50% and categorised as a fence-sitter. The most reliable metric is age because voters are separated into polling lanes based on age. Additionally we have also categorised the 222 Parliament constituencies as rural, semi-urban or urban based on satellite imagery (reference). The descriptions of each category are:
Rural = villages (kampungs) / small towns / farmland distributed within the seat. Rural seats tend to be physically large with a low population.
Semi-urban = larger towns and/or numerous small towns, may include villages as well
Urban = cities where a majority of the seat is covered by some form of urban development
For this report we will focus on how Pakatan Rakyat (PR) and Barisan Nasional (BN) performed with regular Malay voters (pengundi biasa) in Peninsular Malaysia. 184 of the total 222 Parliament seats are in Peninsular Malaysia, where most of the Malay electorate is concentrated.
Elections are won based on the number of seats. However our analysis will mainly be on the Malay electorate treated as a set of voters ignoring constituency boundaries. We will examine this at the state-level and for Peninsular Malaysia. This will allow us to see patterns that are not obvious at the seat-level.
Postal and early voters are not part of this analysis. They need to be analysed separately due to their different voting process and difficulties in campaigning to both groups.
Please remember that unless otherwise stated, all statistics in this analysis refer to regular voters in Peninsular Malaysia only.
2. The Ethnic Divide
The 2 graphs below show the gap between ethnic groups in Malaysia by age in 1970 and 2010, from ages 0 – 39 years. This was taken from the National Census.
The top-most line (red) represents Malays, followed by Chinese (yellow), Other Bumiputera (purple) and Indians (blue). Both graphs use the same scale. Compared to 1970 there are more Malay youth in 2010 relative to youth of other ethnicities.
In 1970, the division between races within the total population (10.8 million) was:
- 44.32% Malay
- 34.34% Chinese
- 8.99% Indian
- 11.69% Bumiputera (non-Malay)
- 0.67% Others
In 2010, the division between races within the total population (28.9 million) was:
- 55.07% Malay
- 24.34% Chinese
- 7.35% Indian
- 11.94% Bumiputera (non-Malay)
- 1.3% Others
In GE13 (2013), the division between races among total voters is estimated to be:
- 52.63% Malay
- 29.68% Chinese
- 7.31% Indian
- 8.96% Bumiputera (non-Malay)
- 1.43% Others
It is clear that there is an increasing gap between the Malay electorate and other ethnic groups in terms of population. Assuming that voter registration rates remains the same across ethnic groups, Malay voters will be the deciding force in future elections.
3. Summary of GE 13 Results
In GE13 PR won 89 seats while BN won 133 seats. In Peninsular Malaysia PR won 80 seats while BN won 85 seats. The breakdown of seats won by urban development category is shown in the table below:
| Urban Development Category / Party | DAP | PKR | PAS | BN | Total |
| Rural | 2 | 2 | 11 | 66 | 81 |
| Semi-Urban | 10 | 14 | 6 | 14 | 44 |
| Urban | 19 | 12 | 4 | 5 | 40 |
In chart form:
Focusing on the 3 main ethnic groups, the number of regular voters of each ethnicity in seats won by respective parties is shown in the table below:
| Ethnic Group | DAP | PKR | PAS | BN | Total |
| Malay | 601356 | 1217460 | 1346233 | 3261493 | 6426542 |
| Chinese | 1484498 | 825648 | 206059 | 900593 | 3416798 |
| Indian | 289325 | 285472 | 92712 | 290749 | 958258 |
In chart form:
By comparing the two sets of data it is clear that BN won the most rural seats and have the most Malay voters (50.75% of the Peninsular Malaysia total).
This table shows the number of ethnic majority seats won by each party:
| Ethnic Majority | DAP | PKR | PAS | BN | Total |
| Malay majority | 0 | 17 | 20 | 77 | 114 |
| Chinese majority | 22 | 0 | 0 | 0 | 22 |
| Mixed | 9 | 11 | 1 | 8 | 29 |
In chart form:
Some key points from this chart:
- DAP won all the Chinese majority seats
- PKR won a mix of Malay majority and Mixed seats
- PAS victories were almost entirely in Malay majority seats
- BN victories were mainly in Malay majority seats
The chart below shows the distribution of voters by ethnicity in Peninsular Malaysia:
The following series of charts show the distribution of voters of each ethnic group based on seats won by each party. For example, this chart shows that 51% of Malay voters reside in seats won by BN:
This table shows the total number of Malay voters by urban development category in seats won by each party:
| Ethnic Group | DAP | PKR | PAS | BN | PR | BN(%) | PR(%) | Ethnic Total |
| Rural Malay | 42604 | 99888 | 713951 | 2407336 | 856443 | 73.76 | 26.24 | 3263779 |
| Semi-Urban Malay | 208735 | 602279 | 367380 | 685766 | 1178394 | 36.79 | 63.21 | 1864160 |
| Urban Malay | 350017 | 515293 | 264902 | 168391 | 1130212 | 12.97 | 87.03 | 1298603 |
Key points for this section are:
- DAP and PKR victories were won mainly in urban and semi-urban seats.
- PR parties represent 74% of Chinese voters; 68% if PAS is not included
- BN represents a slim majority of Malay voters at 51% of the Peninsular Malaysia total
- BN won 83.5% of rural seats containing 73.76% of rural Malay voters
- PR parties won seats containing 63.21% of semi-urban and 87.03% of urban Malay voters
- PKR represents the most number of urban Malay voters
- BN represents more semi-urban Malay voters than any component PR party
- BN represents 51% of the Malay electorate while PAS represents 21%
- 70% of the Malay voters in PAS seats are in Kelantan and Terengganu
This section is a general overview that shows the size of the electorate represented by each party. However representation is not the same as support, as each seat has supporters of both PR and BN. Support is covered in the following sections.
4. Measuring Regular Voter Support by Age
During the election voters are grouped into polling lanes based on age. The results of each lane are reflective of support by the age group. By mapping registered voters to polling lane results we can estimate the support by race and age.
The scatter plot below shows the average probability of Malay voters in a specific polling lane voting for PR:
Each point in the graph represents one polling lane. The horizontal scale shows the percentage of Malay voters in the lane. The vertical scale shows the percentage probability of Malay voters in that lane of voting for PR. Fence-sitters are not shown in this graph.
Taking the highlighted point above – for that specific polling lane with 30.24% Malay voters, there is a 75.69% chance that those Malay voters will vote for PR. The remaining 69.76% of voters of other ethnicities will have their own set of probabilities.
Any point above the 50% probability line is good for PR. From this scatter plot we can tell that PR does well in mixed polling lanes. When the percentage of Malay voters is between 10% – 50% support is clustered around the 50% – 70% range.
When the percentage of Malay voters increases over 44% the Malay vote begins to split. BN has better support with 65% of polling lanes having less than 50% probability of voting for PR. Including fence-sitters, BN does better in 44.5% of polling lanes.
For comparison, here is the scatter plot for Chinese voters:
There is a clear trend of the probability of Chinese voters voting for PR increasing as the percentage of Chinese voters increase.
The following graph shows how average Malay voter support shifted by age group, divided by seats won by respective parties. This was calculated by taking the average of individual support values from voters across all seats for each age group. This gives a picture of how Malay voters in those seats favoured the winning party.
Some key points from this graph are:
- DAP had the highest positive swing from Malays of almost all ages
- PAS had positive swing from voters below 36 years, with negative swing from older voters
- PR had positive swing from voters below 36 years in seats won by BN, with negative swing from older voters
- PKR had the least improvement with positive swing from voters below 30 years and a much higher negative swing from older voters
- The common pattern among all parties is that the youth swung closer to PR while middle-aged and older Malay voters showed less improvement or swung away from PR
This graph shows the average level of support from Malay voters in seats won by respective parties after the swing in GE13. Anything close to or above 50% is good for PR.
Some key points from this graph are:
- PAS has the highest level of support from Malay voters in their seats
- PKR has marginal (close to 50%) support from Malay voters in their 40s and below
- DAP has weak support from Malay voters in their 40s and below
- PR has the weakest support from Malay voters in seats won by BN
- PR would need support to swing by an average of 10% in BN seats to secure the middle-aged and younger Malay vote
- There is a downward trend in support from Malay voters as they get older, though PAS remains in the best position
This section describes voter support by age, taken as an average of support across voters within each age group. These graphs do not show the number of voters – a large negative swing from a small group of voters can be easily offset by a small positive swing from a large group of voters.
These graphs only serve as an indicator of where average support levels are at and what direction they are heading in. Because averages of positive and negative swing values can results in a net positive or negative, we need to look at individual voter support levels to get a more accurate reading.
5. Measuring Overall Regular Voter Support
Because individual voters have their own support and swing statistics, we can calculate the proportion of the electorate that is leaning towards PR (>50% probability of voting PR); leaning towards BN (<50% probability of voting PR); and on the fence (50% probability of voting PR). Voters who are leaning BN might still vote for PR, so this metric is only an indicator of how good the odds are for PR/BN to win support from the electorate.
The two charts below show the predicted and estimated support from all voters in Peninsular Malaysia. By comparing the two we can see how support has shifted.
Prior to GE13, the electorate was evenly divided between PR and BN with 20% being fence-sitters. After GE13 PR increased their share of support by 4% while BN lost 4%. The proportion of fence-sitters remains the same at 20%.
The charts below show voter support from the Malay electorate:
Based on GE12 results, the predicted support for PR from the Malay electorate was not high at 32%. After GE13 support for PR dropped to 30%. It is worth noting that 59% of the Malay electorate are leaning BN but BN seats only represent 51% of the Malay electorate.
In the previous section we showed there was a positive swing to PR from young Malay voters. When we measure total support from Malay voters aged 35 and below (in 2013) we get the following charts:
PR gained 5% support from the Malay youth while BN lost 3% support. The percentage of fence-sitters reduced by 2%.
The table below show additional insight into differences between rural and urban Malay voters, for both total voters and the overall youth. The estimated GE13 support is shown, with swing values in parentheses.
Total Malay Electorate
| Urban Development Category | Leaning PR | Fence-sitters | Leaning BN |
| Rural | 28% (-3%) | 6% (-1%) | 66% (+4%) |
| Semi-Urban | 30% (-1%) | 13% (no change) | 57% (+1%) |
| Urban | 38% (+3%) | 17% (-4%) | 45% (+1%) |
Malay Youth (<=35 years old)
| Urban Development Category | Leaning PR | Fence-sitters | Leaning BN |
| Rural | 35% (no change) | 5% (no change) | 60% (no change) |
| Semi-Urban | 40% (+7%) | 10% (no change) | 50% (-7%) |
| Urban | 48% (+13%) | 9% (-7%) | 43% (-6%) |
Malay Youth (<=35 years old), in seats won
| Winning Party | Leaning PR | Fence-sitters | Leaning BN |
| PKR | 45% (+6%) | 6% (-8%) | 49% (+2%) |
| DAP | 35% (+18%) | 22% (-4%) | 43% (-14%) |
| PAS | 73% (+8%) | 5% (+1%) | 22% (-9%) |
| BN | 22% (no change) | 6% (no change) | 72% (no change) |
When we look at the total Malay electorate, BN gained support of 4% in rural seats. This is not a good indicator for PR as 73/81 rural seats are Malay majority seats.
PR made large gains from Malay youth in urban seats. However the swing in support from Malay youth in rural seats did not result in any significant change. While PR (primarily PAS) did make gains in their rural seats, any gains were offset by BN’s victories in their rural seats.
PAS has the highest support from Malay youth in seats that they won, while DAP showed the most improvement. However the number of Malay youths in DAP seats is small at 205 thousand versus PKR’s 413 thousand, PAS’ 516 thousand and BN’s 1.14 million.
Key points for this section:
- PR gained support of 4% of the total Peninsular Malaysia electorate while BN lost 4%
- PR lost support of 2% from the total Malay electorate while BN gained 2%
- PR made gains from urban Malay voters but lost support from rural and semi-urban voters
- BN increased support by 1% in semi-urban and urban seats but gained 4% support in rural seats
- PR gained support of 5% from Malay youth while BN lost 3%. This change in support came from semi-urban and urban seats
- There was no significant change in support from Malay youth in rural seats
- PAS had the highest support from Malay youth
- 59% of the Malay electorate are leaning BN but BN seats only represent 51% of the Malay electorate
6. Performance by State
In this section we will present a series of notes and statistics covering Malay voter support in each state.
6.1 Overall support and swing values from Malay voters
The estimated GE13 support is shown, with swing values in parentheses. The number of seats in each state is shown in parentheses.
| State | Leaning PR | Fence-sitters | Leaning BN |
| Johor | 10% (+9%) | 17% (+3%) | 73% (-12%) |
| Kedah | 36% (-16%) | 7% (+1%) | 57% (+15%) |
| Kelantan | 62% (-6%) | 4% (no change) | 34% (+6%) |
| Melaka | 10% (+8%) | 10% (-1%) | 80% (-7%) |
| Negeri Sembilan | 8% (no change) | 13% (no change) | 79% (no change) |
| Pahang | 15% (-1%) | 7% (-2%) | 78% (+3%) |
| Pulau Pinang | 32% (-4%) | 20% (-2%) | 48% (+6%) |
| Perak | 18% (-9%) | 14% (-2%) | 68% (+11%) |
| Perlis | 20% (-2%) | 3% (+1%) | 77% (+1%) |
| Selangor | 37% (no change) | 13% (-3%) | 50% (+3%) |
| Terengganu | 42% (+12%) | 5% (+1%) | 53% (-13%) |
| WP Kuala Lumpur | 28% (-10%) | 21% (-7%) | 51% (+17%) |
| WP Putrajaya | 0% (-10%) | 0% (-6%) | 100% (+16%) |
Putrajaya is notable for having 100% of its Malay voters leaning towards BN. The number of voters more than doubled since GE12, growing from 6,608 voters (93% Malay) to 15,791 voters (94% Malay). 46% of the new voters were transferred-in from other states.
In GE13 Husam Musa (PAS) received 31% of the vote. In simple terms this means that there is a 31% probability of voters voting for PR in Putrajaya. For a 94% Malay majority seat that translates into Malay voters considered to be leaning BN due to the voters having a <50% chance of voting for PR.
6.2 Swing and average support levels from Malay voters by age
The number of seats won by each party is listed after each state. Support levels in the note refer to average support by age group for specific parties. Swing values are calculated from averages.
For each state we will mention if there is a downward trend pattern in swing levels. This is where we observed a tendency for negative swing increasing with age or positive swing decreasing with age.
| State | Note |
| Johor
PKR: 1 DAP: 4 BN: 21 |
Positive swing in support for PR in all seats, starting as high as +9% (PKR) with the young voters and declining with older voters.
PKR seats had the highest level of support (44% and below) followed by DAP (40% and below) and BN (31% and below).
Downward trend observed.
|
| Kedah
PKR: 4 PAS: 1 BN: 10 |
Negative swing in support for PR in PAS, PKR and BN seats, between -1% to -8%. PAS had less negative swing compared to PKR and BN.
PAS seats had the highest level of support (55% and below) followed by PKR seats (50% and below) and BN seats (47% and below).
Downward trend observed.
|
| Kelantan
PAS: 9 BN: 5 |
Positive swing in support for PR in PAS seats from voters aged 30 and below. Negative swing for other ages in PAS seats. Negative swing in BN seats for all ages.
Support in PAS seats remained high at 62% and below. Support for PR in BN seats was at 48% and below.
Downward trend observed.
|
| Melaka
PKR: 1 DAP: 1 BN: 4 |
Positive swing in support for PR in all seats, except for PKR between ages 42 – 54 where it was negative.
PKR seats had the highest support (43% and below) while DAP maintained a range of support between 34% – 39% with support being higher for middle-aged voters aged 36 – 45 (40% – 43% support).
Support remained low in BN seats at 34% and below.
Downward trend observed.
|
| Negeri Sembilan
PKR: 1 DAP: 2 BN: 5 |
Positive swing in support for PR in DAP and BN seats for ages 37 and below and 46 – 51. Negative swing for other age groups. Negative swing for PKR seats in all age groups.
PKR seats had the highest support (43% and below) with support being highest among middle-aged voters (36 – 41). DAP seats had support at 39% and below while BN seats had support at 35% and below.
Downward trend observed. |
| Pahang
PKR: 2 DAP: 1 PAS: 1 BN: 10 |
Positive swing in support for PR in all seats for ages 36 and below, negative swing for other ages.
Support was highest in PKR seats (53% and below) followed by PAS seats (50% and below). DAP and BN seats had very similar levels of support between 36% – 38%.
Downward trend observed.
|
| Pulau Pinang
PKR: 3 DAP: 7 BN: 3 |
Positive swing in support for PR in DAP seats for all ages except 41 – 56 where swing was negative. Negative swing in PKR seats for all ages.
Positive swing in BN seats for specific age ranges: 25 and below, 36 – 39, 68 and above. There was negative swing in other age groups.
Support was highest in PKR seats at 56% and below, with highest levels between ages 35 – 40. Support in DAP seats was at 49% and below. Support in BN seats was at 44% and below, with highest levels between ages 37 – 42.
Downward trend observed for ages 37 – 56.
|
| Perak
PKR: 3 DAP: 7 PAS: 2 BN: 12 |
Positive swing in support for PR in DAP seats for ages 37 and below. Positive swing in PKR seats for ages 28 and below. Negative swing in PAS and BN seats for all ages.
Support was highest in PAS seats (52% and below) followed by PKR seats (49% and below), DAP seats (46% and below) and BN seats (42% and below).
|
| Perlis
BN: 3 |
Positive swing in support for PR in BN seats for all ages except 65 and 82 year olds. The swing was highest among the youth at +7.4% and below.
Support in BN seats remained at 45% and below.
Downward trend observed.
|
| Selangor
PKR: 9 DAP: 4 PAS: 4 BN: 5 |
Positive swing in support for PR in DAP seats for all ages. Positive swing in PKR seats for voters aged 35 and below, negative swing for other ages.
Positive swing in PAS seats for voters aged 33 – 36, negative swing for other ages. Negative swing in BN seats for all ages except 25 – 26 where there was a marginal <1% positive swing.
Support was highest in PAS seats at 54% and below, with the highest levels in the middle-aged group (38 – 45).
DAP seats maintained similar support levels as PAS with support at 53% and below. PKR seats had support at 51% and below.
BN seats had support at 45% and below.
Downward trend observed.
|
| Terengganu
PAS: 4 BN: 4 |
Positive swing in support for PR in all seats, with swing being highest with the younger generation at +8% and below.
Support remained high in PAS seats at 57% and below, while BN seats had support at 44% and below.
Downward trend observed.
|
| WP Kuala Lumpur
PKR: 4 DAP: 5 BN: 2 |
Positive swing in support for PR in seats won by DAP, starting high at +8% among youth and declining with age.
Negative swing in support for PR in PKR and BN seats apart from voters aged 27 – 29 where support increased.
Support was highest in DAP seats at 55% and below. Support for PR in PKR seats was highest in the middle-aged group (38 – 45) at just above 50%.
Support in BN seats closely mirrored PKR with BN seats being 1-2% lower for each age group.
Downward trend observed.
|
| WP Putrajaya
BN: 1 |
Positive swing from voters age 29 and below. Negative swing for voters of other ages.
Support for PR was at its highest for 27 year olds (37%) with decreasing values for other ages.
Downward trend observed.
|
Perlis recorded an average swing towards PR among Malay voters of all ages, yet PR recorded a net loss in terms of support from the Malay electorate (refer to the previous section). This is because of the effect of averages, where voters with high positive swing values can hide the impact of voters with low negative swing values.
Voter population by age is another factor that can affect the results as a large average swing to PR from a smaller population is over-ridden by a low average positive swing.
7. Summary
The next general election will not be decided by Malay voters alone – obviously the way constituency boundaries are structured will add weight to non-Malay voters. Postal voters will also be a factor on a per-seat basis.
In preparing this report we hope to highlight how important the Malay voter base will be in future elections and how PR and BN has performed so far.
Some points worth noting are:
7.1 The number of Malay-majority seats will increase with time
GE14 and GE15 will see an increasing number of Malay-majority seats and mixed seats. Between GE12 and GE13 138/165 Peninsular Malaysia seats had an increase in the percentage of Malay voters.
In GE12 there were 26 Chinese-majority seats in Peninsular Malaysia. In GE13 there were 22. Serdang (P102), Rasah (P130), Kluang (P152) and Taiping (P60) have become mixed seats when previously they were Chinese-majority seats.
Lumut (P74) has become a Malay-majority seat when previously it was mixed. Raub (P80) is close to becoming a Malay-majority seat with 49.8% Malay voters.
The shift is already taking place. Parties/politicians that rely on the non-Malay vote will face difficulties unless they change their strategy.
7.2 BN made significant gains among Malay electorate
For state-level support, BN also made gains with Malay voters in Kedah (+15%), Kelantan (+6%), Penang (+6%), Perak (+11%) and WP Kuala Lumpur (+17%). BN lost support in Terengganu (-13%), Johor (-12%) and Melaka (-7%). In GE13 BN gained control of the state government in Kedah and Perak so this swing was quite significant.
7.3 PAS performed well with Malay voters
Within PR, PAS contested and won seats with the most number of Malay voters. In Kelantan and Terengganu they won 14 seats and retained a high percentage of support.
Within Peninsular Malaysia PAS has the highest support from Malay youth, with 73% of Malay youth in PAS seats are leaning towards PAS (with +8% swing in GE13).
BN is close with support from 72% of Malay youth in their seats, but saw no improvement in GE13.
In Johor (53% Malay voters, 17/26 Malay-majority seats) and Melaka (58% Malay voters, 5/6 Malay-majority seats) BN still retains a majority of support in each state.
However in Terengganu (96% Malay voters) BN only has support from 53% of the electorate. PAS made gains from voters in all Terengganu seats in GE13 so there is a chance for it to win the majority of Malay support in GE14. This can translate to control of the state government.
PAS made significant gains and maintained high average support levels from Malay voters in their seats. BN still retains an advantage in their rural seats. PAS’ gains in semi-urban and urban areas could be an indicator of future success in these areas.
7.4 Young Malay voters swung to PR, but gains were marginal in rural seats
In this analysis there is a common pattern that we observed of young Malays moving towards PR and middle-aged and older Malays moving away from PR or showing less improvement. The downward trend was observed in every state, with some trends starting with the middle-aged voters instead of young voters.
PR made gains from urban Malay voters and young Malay voters in semi-urban and urban seats. However they lost support from Malay voters in rural seats.
Despite young Malay voters in rural seats swinging to PR, on a Peninsular Malaysia-level the shift was not enough to increase the share of these young voters leaning towards PR.
7.5 PR needs to take advantage of gains with young Malay voters
If PR was still united, it would be possible to win more Malay-majority urban and semi-urban seats depending on the following factors:
- The number of young Malays joining the electorate. Significant numbers of young Malays can help offset the expected lower support from older-generation Malay voters
- Repeating the success of GE13 in terms of % swing of support
- Finding ways to gain support from older-generation Malay voters will reduce dependency on non-Malay voters
- Avoiding 3-corner fights which could split the Malay vote
7.6 Future of the Opposition
PKR and DAP have not had as much success winning rural seats compared to PAS and BN. Without PAS continuing to represent PR in those seats, a new coalition will have difficulties winning GE14.
The possibility of 3-corner fights between PAS, BN and Gerakan Harapan Baru (GHB) increases the likelihood of BN winning seats with <50% of the vote. This could happen even in Kelantan and Terengganu if enough PAS leaders and members leave to join the new party.
